S a M P L E Grand Rounds Presentation Pachyonychia

S A M P L E

Grand Rounds Presentation

Pachyonychia Congenita

presented by

[Name, Credentials, Affiliation]

sponsored by PC Project

Table of Contents

Welcome—PC Project, IPCRR, IPCC ...... 2 IPCRR Status Data / Invitation to join the IPCC Chart of Mutations / Schematic of Mutations ...... 3 IPCRR Data Summary (712 genetically confirmed with PC) ...... 4 Presentation of Patients for Grand Rounds PC-K6a IPCRR #72 K6a L468P ...... 5 IPCRR #344 K6a L469P ...... 6 IPCRR #820 K6a L469R ...... 7 IPCRR #1338 K6a N172del ...... 8 PC-K6b IPCRR #641 K6b E472K ...... 9 PC-K6c IPCRR #713 K6c E472K ...... 10 PC-K16 IPCRR #35 K16 R127P ...... 11 IPCRR #66 K16 L132P ...... 12 IPCRR #850 K16 R418P ...... 13 IPCRR #1578 K16 M121T ...... 14 PC-K17 IPCRR #515 K17 N92S ...... 15 List of Selected Articles ...... 16 Full text articles ......

Sep 2016 Welcome! Pachyonychia Congenita Project (PC Project) is a public charity in the USA which sponsors the International Pachyonychia Congenita Research Registry (IPCRR) serving patients around the world. Beginning with 3 patients in 2004, the IRB- approved registry has now enrolled over 1700 individuals diagnosed with PC. Of these, 968 have complete data on file in the IPCRR including patient input- physician validated data, images and genetic testing results and of those 727 have confirmed PC.

Location of genetically confirmed individuals with PC Pachyonychia Congenita Project (PC Project) also sponsors the International Pach- yonychia Consortium (IPCC) a collaboration of physicians and scientists working in clinical care as well as basic and translational research to improve lives for those with PC and tofind effective treatments. If you’d like to join this open membership group and receive the quarterly newsletters, please include your name and contact details in an email request to [email protected].

3 International Pachyonychia Congenita Research Registry (IPCRR) Questionnaire Summary Data as of July 2016

IPCRR Data PC-K6a PC-K6b PC-K6c PC-K16 PC-K17 All PC Genes 7 Jul 2016 N=276 N=67 N=22 N=231 N=116 N=712

Toenails Dystrophy 273 of 276 (99%) 66 of 67 (99%) 13 of 22 (59%) 223 of 231 (97%) 112 of 116 (97%) 687 of 712 (96%) all 10 toenails thickened 259 of 276 (94%) 24 of 67 (36%) 0 of 22 (00%) 94 of 231 (41%) 87 of 116 (75%) 464 of 712 (65%) 7-9 toenails thickened 8 of 276 (03%) 14 of 67 (21%) 0 of 22 (00%) 29 of 231 (13%) 10 of 116 (09%) 61 of 712 (09%) 4-6 toenails thickened 2 of 276 (01%) 19 of 67 (28%) 3 of 22 (14%) 59 of 231 (26%) 7 of 116 (06%) 90 of 712 (13%) 1-3 toenails thickened 4 of 276 (01%) 9 of 67 (13%) 10 of 22 (45%) 41 of 231 (18%) 8 of 116 (07%) 72 of 712 (10%) Toenails - Onset N = 273 N = 66 N = 13 N = 223 N = 112 N = 696 Birth or less than 1 year 239 of 273 (88%) 9 of 66 (14%) 1 of 13 (08%) 44 of 223 (20%) 79 of 112 (71%) 372 of 696 (53%) 1 to 4 years old 30 of 273 (11%) 16 of 66 (24%) 6 of 13 (46%) 72 of 223 (32%) 24 of 112 (21%) 148 of 696 (21%) 5 to 14 years old 4 of 273 (01%) 34 of 66 (52%) 4 of 13 (31%) 68 of 223 (30%) 9 of 112 (08%) 119 of 696 (17%) 15 years and over 0 of 273 (00%) 7 of 66 (11%) 2 of 13 (15%) 38 of 223 (17%) 1 of 112 (01%) 48 of 696 (07%) Fingernail Dystrophy 273 of 276 (99%) 31 of 67 (46%) 0 of 22 (00%) 136 of 231 (59%) 98 of 116 (84%) 538 of 712 (76%) all 10 fingernails thickened 244 of 276 (88%) 4 of 67 (06%) 0 of 22 (00%) 71 of 231 (31%) 50 of 116 (43%) 369 of 712 (52%) 7-9 fingernails thickened 9 of 276 (03%) 4 of 67 (06%) 0 of 22 (00%) 11 of 231 (05%) 12 of 116 (10%) 36 of 712 (05%) 4-6 fingernails thickened 14 of 276 (05%) 16 of 67 (24%) 0 of 22 (00%) 28 of 231 (12%) 27 of 116 (23%) 85 of 712 (12%) 1-3 fingernails thickened 6 of 276 (02%) 7 of 67 (10%) 0 of 22 (00%) 26 of 231 (11%) 9 of 116 (08%) 48 of 712 (07%) Fingernails - Onset N = 273 N = 31 N = 0 N = 136 N = 98 N = 538 Birth or less than 1 year 242 of 273 (89%) 4 of 31 (13%) 0 of 0 (00%) 33 of 136 (24%) 72 of 98 (73%) 351 of 538 (65%) 1 to 4 years old 27 of 273 (10%) 7 of 31 (23%) 0 of 0 (00%) 40 of 136 (29%) 19 of 98 (19%) 93 of 538 (17%) 5 to 14 years old 3 of 273 (01%) 11 of 31 (35%) 0 of 0 (00%) 33 of 136 (24%) 5 of 98 (05%) 52 of 538 (10%) 15 years and over 1 of 273 (00%) 10 of 31 (32%) 0 of 0 (00%) 31 of 136 (23%) 3 of 98 (03%) 45 of 538 (08%) Plantar Keratoderma 243 of 276 (88%) 64 of 67 (96%) 19 of 22 (86%) 227 of 231 (98%) 91 of 116 (78%) 644 of 712 (90%) Always (never goes away) 233 of 276 (84%) 63 of 67 (94%) 19 of 22 (86%) 226 of 231 (98%) 74 of 116 (64%) 615 of 712 (86%) Sometimes (clear up at times) 5 of 276 (02%) 1 of 67 (01%) 0 of 22 (00%) 1 of 231 (00%) 13 of 116 (11%) 20 of 712 (03%) Seldom (feet usually clear) 6 of 276 (02%) 0 of 67 (00%) 0 of 22 (00%) 0 of 231 (00%) 4 of 116 (03%) 10 of 712 (01%) Plantar Keratoderma- Onset N = 243 N = 64 N = 19 N = 227 N = 91 N = 644 Birth or less than 1 year 35 of 243 (14%) 2 of 64 (03%) 1 of 19 (05%) 20 of 227 (09%) 12 of 91 (13%) 70 of 644 (11%) 1 to 4 years old 143 of 243 (59%) 20 of 64 (31%) 9 of 19 (47%) 124 of 227 (55%) 29 of 91 (32%) 325 of 644 (50%) 5 to 14 years old 62 of 243 (26%) 41 of 64 (64%) 9 of 19 (47%) 77 of 227 (34%) 37 of 91 (41%) 226 of 644 (35%) 15 years and over 3 of 243 (01%) 1 of 64 (02%) 0 of 19 (00%) 7 of 227 (03%) 14 of 91 (15%) 25 of 644 (04%) Plantar Pain 236 of 243 (97%) 64 of 64 (100%) 19 of 19 (100%) 217 of 227 (96%) 77 of 91 (85%) 613 of 644 (95%) Often require meds for pain 61 of 243 (25%) 11 of 64 (17%) 5 of 19 (26%) 70 of 227 (31%) 17 of 91 (19%) 164 of 644 (25%) Very painful, but do not use meds 105 of 243 (43%) 32 of 64 (50%) 11 of 19 (58%) 103 of 227 (45%) 28 of 91 (31%) 279 of 644 (43%) Somewhat painful 70 of 243 (29%) 21 of 64 (33%) 3 of 19 (16%) 47 of 227 (21%) 32 of 91 (35%) 173 of 644 (27%) Palmar Keratoderma 152 of 276 (55%) 28 of 67 (42%) 6 of 22 (27%) 173 of 231 (75%) 60 of 116 (52%) 419 of 712 (59%) Always (never goes away) 81 of 276 (29%) 11 of 67 (16%) 2 of 22 (09%) 139 of 231 (60%) 16 of 116 (14%) 249 of 712 (35%) Sometimes (clear up at times) 26 of 276 (09%) 6 of 67 (09%) 1 of 22 (05%) 14 of 231 (06%) 19 of 116 (16%) 66 of 712 (09%) Seldom (hands usually clear) 45 of 276 (16%) 11 of 67 (16%) 3 of 22 (14%) 20 of 231 (09%) 24 of 116 (21%) 103 of 712 (14%) Other Oral Leukokeratosis 247 of 276 (89%) 18 of 67 (27%) 4 of 22 (18%) 83 of 231 (36%) 29 of 116 (25%) 381 of 712 (54%) Cysts 178 of 276 (64%) 47 of 67 (70%) 4 of 22 (18%) 59 of 231 (26%) 107 of 116 (92%) 395 of 712 (55%) Follicular Hyperkeratosis 155 of 276 (56%) 29 of 67 (43%) 0 of 22 (00%) 28 of 231 (12%) 75 of 116 (65%) 287 of 712 (40%) Natal or Prenatal Teeth 8 of 276 (03%) 0 of 67 (00%) 0 of 22 (00%) 0 of 231 (00%) 88 of 116 (76%) 96 of 712 (13%)

K6a L468P PC-K6a Data for IPCRR #72 N=276

Toenails Dystrophy 273 of 276 (99%) all 10 toenails thickened 259 of 276 (94%) 7-9 toenails thickened 8 of 276 (03%) 4-6 toenails thickened 2 of 276 (01%) 1-3 toenails thickened 4 of 276 (01%) Toenails - Onset N = 273 Birth or less than 1 year 239 of 273 (88%) 1 to 4 years old 30 of 273 (11%) 5 to 14 years old 4 of 273 (01%) 15 years and over 0 of 273 (00%) Fingernail Dystrophy 273 of 276 (99%) all 10 fingernails thickened 244 of 276 (88%) 7-9 fingernails thickened 9 of 276 (03%) 4-6 fingernails thickened 14 of 276 (05%) 1-3 fingernails thickened 6 of 276 (02%) Fingernails - Onset N = 273 Birth or less than 1 year 242 of 273 (89%) 1 to 4 years old 27 of 273 (10%) 5 to 14 years old 3 of 273 (01%) 15 years and over 1 of 273 (00%) Plantar Keratoderma 243 of 276 (88%) Always (never goes away) 233 of 276 (84%) Sometimes (clear up at times) 5 of 276 (02%) Seldom (feet usually clear) 6 of 276 (02%) Plantar Keratoderma- Onset N = 243 Birth or less than 1 year 35 of 243 (14%) 1 to 4 years old 143 of 243 (59%) 5 to 14 years old 62 of 243 (26%) 15 years and over 3 of 243 (01%) Plantar Pain 236 of 243 (97%) Often require meds for pain 61 of 243 (25%) Very painful, but do not use meds 105 of 243 (43%) Somewhat painful 70 of 243 (29%) Palmar Keratoderma 152 of 276 (55%) Always (never goes away) 81 of 276 (29%) Sometimes (clear up at times) 26 of 276 (09%) Seldom (hands usually clear) 45 of 276 (16%) Other Oral Leukokeratosis 247 of 276 (89%) Cysts 178 of 276 (64%) Follicular Hyperkeratosis 155 of 276 (56%) Natal or Prenatal Teeth 8 of 276 (03%)

K6a L469P PC-K6a Data for IPCRR #344 N=276

K6a L469R PC-K6a Data for IPCRR #820 N=276

K6a N172del PC-K6a Data for IPCRR #1338 N=276

K6b E472K PC-K6b Data for IPCRR #641 N=67

Toenails Dystrophy 66 of 67 (99%) all 10 toenails thickened 24 of 67 (36%) 7-9 toenails thickened 14 of 67 (21%) 4-6 toenails thickened 19 of 67 (28%) 1-3 toenails thickened 9 of 67 (13%) Toenails - Onset N = 66 Birth or less than 1 year 9 of 66 (14%) 1 to 4 years old 16 of 66 (24%) 5 to 14 years old 34 of 66 (52%) 15 years and over 7 of 66 (11%) Fingernail Dystrophy 31 of 67 (46%) all 10 fingernails thickened 4 of 67 (06%) 7-9 fingernails thickened 4 of 67 (06%) 4-6 fingernails thickened 16 of 67 (24%) 1-3 fingernails thickened 7 of 67 (10%) Fingernails - Onset N = 31 Birth or less than 1 year 4 of 31 (13%) 1 to 4 years old 7 of 31 (23%) 5 to 14 years old 11 of 31 (35%) 15 years and over 10 of 31 (32%) Plantar Keratoderma 64 of 67 (96%) Always (never goes away) 63 of 67 (94%) Sometimes (clear up at times) 1 of 67 (01%) Seldom (feet usually clear) 0 of 67 (00%) Plantar Keratoderma- Onset N = 64 Birth or less than 1 year 2 of 64 (03%) 1 to 4 years old 20 of 64 (31%) 5 to 14 years old 41 of 64 (64%) 15 years and over 1 of 64 (02%) Plantar Pain 64 of 64 (100%) Often require meds for pain 11 of 64 (17%) Very painful, but do not use meds 32 of 64 (50%) Somewhat painful 21 of 64 (33%) Palmar Keratoderma 28 of 67 (42%) Always (never goes away) 11 of 67 (16%) Sometimes (clear up at times) 6 of 67 (09%) Seldom (hands usually clear) 11 of 67 (16%) Other Oral Leukokeratosis 18 of 67 (27%) Cysts 47 of 67 (70%) Follicular Hyperkeratosis 29 of 67 (43%) Natal or Prenatal Teeth 0 of 67 (00%)

K6c E472K PC-K6c Data for IPCRR #713 N=22

Toenails Dystrophy 13 of 22 (59%) all 10 toenails thickened 0 of 22 (00%) 7-9 toenails thickened 0 of 22 (00%) 4-6 toenails thickened 3 of 22 (14%) 1-3 toenails thickened 10 of 22 (45%) Toenails - Onset N = 13 Birth or less than 1 year 1 of 13 (08%) 1 to 4 years old 6 of 13 (46%) 5 to 14 years old 4 of 13 (31%) 15 years and over 2 of 13 (15%) Fingernail Dystrophy 0 of 22 (00%) all 10 fingernails thickened 0 of 22 (00%) 7-9 fingernails thickened 0 of 22 (00%) 4-6 fingernails thickened 0 of 22 (00%) 1-3 fingernails thickened 0 of 22 (00%) Fingernails - Onset N = 0 Birth or less than 1 year 0 of 0 (00%) 1 to 4 years old 0 of 0 (00%) 5 to 14 years old 0 of 0 (00%) 15 years and over 0 of 0 (00%) Plantar Keratoderma 19 of 22 (86%) Always (never goes away) 19 of 22 (86%) Sometimes (clear up at times) 0 of 22 (00%) Seldom (feet usually clear) 0 of 22 (00%) Plantar Keratoderma- Onset N = 19 Birth or less than 1 year 1 of 19 (05%) 1 to 4 years old 9 of 19 (47%) 5 to 14 years old 9 of 19 (47%) 15 years and over 0 of 19 (00%) Plantar Pain 19 of 19 (100%) Often require meds for pain 5 of 19 (26%) Very painful, but do not use meds 11 of 19 (58%) Somewhat painful 3 of 19 (16%) Palmar Keratoderma 6 of 22 (27%) Always (never goes away) 2 of 22 (09%) Sometimes (clear up at times) 1 of 22 (05%) Seldom (hands usually clear) 3 of 22 (14%) Other Oral Leukokeratosis 4 of 22 (18%) Cysts 4 of 22 (18%) Follicular Hyperkeratosis 0 of 22 (00%) Natal or Prenatal Teeth 0 of 22 (00%)

K16 R127P PC-K16 Data for IPCRR #35 N=231

Toenails Dystrophy 223 of 231 (97%) all 10 toenails thickened 94 of 231 (41%) 7-9 toenails thickened 29 of 231 (13%) 4-6 toenails thickened 59 of 231 (26%) 1-3 toenails thickened 41 of 231 (18%) Toenails - Onset N = 223 Birth or less than 1 year 44 of 223 (20%) 1 to 4 years old 72 of 223 (32%) 5 to 14 years old 68 of 223 (30%) 15 years and over 38 of 223 (17%) Fingernail Dystrophy 136 of 231 (59%) all 10 fingernails thickened 71 of 231 (31%) 7-9 fingernails thickened 11 of 231 (05%) 4-6 fingernails thickened 28 of 231 (12%) 1-3 fingernails thickened 26 of 231 (11%) Fingernails - Onset N = 136 Birth or less than 1 year 33 of 136 (24%) 1 to 4 years old 40 of 136 (29%) 5 to 14 years old 33 of 136 (24%) 15 years and over 31 of 136 (23%) Plantar Keratoderma 227 of 231 (98%) Always (never goes away) 226 of 231 (98%) Sometimes (clear up at times) 1 of 231 (00%) Seldom (feet usually clear) 0 of 231 (00%) Plantar Keratoderma- Onset N = 227 Birth or less than 1 year 20 of 227 (09%) 1 to 4 years old 124 of 227 (55%) 5 to 14 years old 77 of 227 (34%) 15 years and over 7 of 227 (03%) Plantar Pain 217 of 227 (96%) Often require meds for pain 70 of 227 (31%) Very painful, but do not use meds 103 of 227 (45%) Somewhat painful 47 of 227 (21%) Palmar Keratoderma 173 of 231 (75%) Always (never goes away) 139 of 231 (60%) Sometimes (clear up at times) 14 of 231 (06%) Seldom (hands usually clear) 20 of 231 (09%) Other Oral Leukokeratosis 83 of 231 (36%) Cysts 59 of 231 (26%) Follicular Hyperkeratosis 28 of 231 (12%) Natal or Prenatal Teeth 0 of 231 (00%)

K16 L132P PC-K16 Data for IPCRR #66 N=231

K16 R418P PC-K16 Data for IPCRR #850 N=231

K16 M121T PC-K16 Data for IPCRR #1578 N=231

14 Selected Articles

Key articles selected from over 700 full text articles with English translations regarding PC and related research and disorders available on the PC Project website www.pachyonychia.org.

Eliason MJ, Leachman SA, Feng BJ, Schwartz ME, Hansen CD. A review of the clinical phenotype of 254 patients with genetically confirmed pachyonychia congenita. Journal of the American Academy of Dermatology. Oct 2012;67(4):680-686. Goldberg I, Fruchter D, Meilick A, Schwartz ME, Sprecher E. Best treatment practices for pachyonychia congenita. Journal of the European Academy of Dermatology and Venere- ology : JEADV. Jan 30 2013. Gonzalez-Ramos J, Sendagorta-Cudos E, Gonzalez-Lopez G, Mayor-Ibarguren A, Feltes- Ochoa R, Herranz-Pinto P. Efficacy of botulinum toxin in pachyonychia congenita type 1: report of two new cases. Dermatologic therapy. Oct 7 2015. Gruber R, Edlinger M, Kaspar RL, et al. An appraisal of oral retinoids in the treatment of pachyonychia congenita. Journal of the American Academy of Dermatology. Jun 2012;66 (6):e193-199. Leachman SA, Hickerson RP, Schwartz ME, Bullough EE, Hutcherson SL, Boucher KM, Hansen CD. Eliason, MJ, Srivatsa GS, Kornbrust DJ, Smith FJ McLean, WH, Milstone LM, Kaspar RL.First-in-human Mutation-targeted siRNA Phase 1b Trial of an Inher- ited Skin Disorder. Mol Ther 2009 Nov 24. McGrath JA. Pachyonychia congenita: cast in translation. The Journal of investigative dermatology. May 2011;131(5):995. McLean WH, Hansen CD, Eliason MJ, Smith FJ. The phenotypic and molecular genetic features of pachyonychia congenita. The Journal of investigative dermatology. May 2011;131(5):1015-1017. Schwartz ME, Zimmerman GM, Smith FJ, Sprecher E. Pachyonychia Congenita Project: A Partnership of Patient and Medical Professional. J Derm Nurses Association. January/ February 2013;5(1):42-47. Schwartz MD, Hansen CD, Paller A, Smith FJD, Sprecher E. Pachyonychia Congenita Overview. The Dermatologist. 2014:22(4); 26-38. Shah S, Boen M, Kenner-Bell B, Schwartz M, Rademaker A, Paller AS. Pachyonychia Congenita in Pediatric Patients: Natural History, Features, and Impact. JAMA dermatology. Oct 16 2013. Wallis T, Poole CD, Hoggart B. Can skin disease cause neuropathic pain? A study in pachyonychia congenita. Clinical and Experimental Dermatology. Sep 11 2015. Wilson NJ, Leachman SA, Hansen CD, et al. A large mutational study in pachyonychia congenita. The Journal of investigative dermatology. May 2011;131(5):1018-1024. Wilson NJ, O'Toole EA, Milstone LM, Hansen CD, Shepherd AA, Ali-Ashadi E, Schwartz ME, McLean WHI, Sprecher E, Smith JFD. The molecular genetic analysis of the expanding pachyonychia congenita case collection. The British Journal of Dermatology. Aug 2014;171(2):343-355.

ORIGINAL ARTICLE

A review of the clinical phenotype of 254 patients with genetically conﬁrmed pachyonychia congenita

Mark J. Eliason, MD,a Sancy A. Leachman, MD, PhD,a Bing-jian Feng, PhD,a Mary E. Schwartz, AA,b and C. David Hansen, MDa Salt Lake City, Utah

Background: Pachyonychia congenita (PC) is a group of autosomal dominant keratinizing disorders caused by a mutation in one of 4 keratin genes. Previous classiﬁcation schemes have relied on data from case series and case reports. Most patients in these reports were not genetically tested for PC.

Objective: We sought to clarify the prevalence of clinical features associated with PC.

Methods: We surveyed 254 individuals with conﬁrmed keratin mutations regarding their experience with clinical ﬁndings associated with PC. Statistical comparison of the groups by keratin mutation was performed using logistic regression analysis.

Results: Although the onset of clinical symptoms varied considerably among our patients, a diagnostic triad of toenail thickening, plantar keratoderma, and plantar pain was reported by 97% of patients with PC by age 10 years. Plantar pain had the most profound impact on quality of life. Other clinical ﬁndings reported by our patients included ﬁngernail dystrophy, oral leukokeratosis, palmar keratoderma, follicular hyperkeratosis, hyperhidrosis, cysts, hoarseness, and natal teeth. We observed a higher likelihood of oral leukokeratosis in individuals harboring KRT6A mutations, and a strong association of natal teeth and cysts in carriers of a KRT17 mutation. Most keratin subgroups expressed a mixed constellation of findings historically reported as PC-1 and PC-2.

Limitations: Data were obtained through questionnaires, not by direct examination. Patients were self- or physician-referred.

Conclusions: We propose a new classiﬁcation for PC based on the speciﬁc keratin gene affected to help clinicians improve their diagnostic and prognostic accuracy, correct spurious associations, and improve therapeutic development. ( J Am Acad Dermatol 10.1016/j.jaad.2011.12.009.)

Key words: genodermatosis; hyperkeratosis; keratin; keratinizing disorder; keratoderma; pachyonychia congenita.

achyonychia congenita (PC) is a group of Abbreviations used: autosomal dominant disorders caused by a mutation in one of 4 keratin genes: KRT6A, IPCRR: International Pachyonychia Congenita P 1-5 Research Registry KRT6B, KRT16,orKRT17. There are an estimated 6 OR: odds ratio 5000 to 10,000 cases worldwide. The variable clin- PC: pachyonychia congenita ical findings affect a number of ectodermal struc- 2 tures, including nails, skin, teeth, and oral mucosa. Lewandowski,9 whose names constitute the eponym 7 8 Although Muller and Wilson and Cantar are for PC type 1, published the first case series of two credited with describing PC in 1904, Jadassohn and siblings in 1906. Kumer and Loos10 proposed a

From the Department of Dermatology, University of Utaha; and Medicine, Salt Lake City, UT 84132. E-mail: david.hansen@ Pachyonychia Congenita Project.b hsc.utah.edu. Funded by the Pachyonychia Congenita Project. Published online January 19, 2012. Conflicts of interest: None declared. 0190-9622/$36.00 Accepted for publication December 8, 2011. Ó 2011 by the American Academy of Dermatology, Inc. Reprint requests: C. David Hansen, MD, Department of doi:10.1016/j.jaad.2011.12.009 Dermatology, University of Utah, 4A330 School of

1 2 Eliason et al JAM ACAD DERMATOL

clinical classification scheme for PC variants based on Participants in the registry were solicited through their report of a 5-generation family with 23 affected an Internet World Wide Web site designed to educate family members. Classification criteria were devel- patients and physicians about PC (www. oped and refined over subsequent years by authors pachyonychia.org). Referral to the registry was per- who painstakingly reviewed and summarized the mitted through patients, family, physicians, and available literature.11-19 Two clinical subtypes family expansion. To be included in the registry, ultimately emerged, the Jadassohn-Lewandowski each patient completed a detailed questionnaire PC type 1 and the Jackson- and provided information re- Lawler PC type 2. garding whether and to what 20 CAPSULE SUMMARY In 1994, Munro et al extent they were affected by studied a large Jackson- the clinical features of PC. d Pachyonychia congenita (PC), a rare Lawler pedigree and linked Patients were also asked genodermatosis caused by mutations in the first PC gene to chromo- about the age of onset and keratin genes, is currently classified as some 17q12-q21. In 1995, the impact each feature had 3 PC-1 and PC-2 based on clinical features. McLean et al identified the on their quality of life. The first causative mutations in d We report the prevalence of clinical completed questionnaire, keratin genes KRT16 and findings in 254 patients with genetically along with photographs of KRT17. Additional mutations confirmed PC. visible skin and nail changes, were subsequently identified d We propose a new classification system was submitted to the IPCRR. KRT6a KRT6be in and genes based on the specific keratin mutation A telephone consultation was encoding the type II keratins (eg, PC-6a, PC-6b, PC-16, PC-17). then arranged with a derma- that form heteropolymers tologist on the Pachyonychia with type I keratins K16 Congenita Project medical 4,5 and K17. The identification of these mutations advisory board to: (1) clarify any confusing and the advent of clinical genetic testing allowed the or missing information from the questionnaire; classification of PC based on clinical and genetic (2) confirm that the clinical features were consistent criteria. with PC; and (3) provide genetic counseling before Erroneous reports of PC manifestations in patients mutation testing. Genetic testing was provided who did not have PC have been clarified by inves- without charge and was performed in Dr Frances 21 tigators through genetic testing. Large, well- Smith’s laboratory, University of Dundee, College of characterized and mutation-confirmed pedigrees Life Sciences, Division of Molecular Medicine, offer the opportunity to draw valid conclusions Dundee, Scotland. Before being released to pa- 22 regarding genotype-phenotype relationships. tients, the results were confirmed by independent However, even these pedigrees are prone to bias testing of a buccal DNA sample by GeneDx because of shared modifier genes and environments (Gaithersburg, Maryland), a US Clinical Laboratory that might influence the clinical presentation. This Improvement Amendmentsecertified laboratory. report summarizes data collected from 254 patients All participant data included in the analysis were with mutation-verified PC (derived from 147 fami- from patients with a confirmed PC keratin mutation. lies) and, to our knowledge, represents the largest and most comprehensive genotype-phenotype Statistical methods study of PC to date. We performed logistic regression analysis to compare how different PC keratin mutations influence the probability of developing a specific clinical METHODS finding. For outcomes such as age of onset and In 2004, the International Pachyonychia quality of life, ordinal logistic regression was used. Congenita Research Registry (IPCRR) was estab- Because there were more KRT6A carriers than other lished by the nonprofit organization Pachyonychia mutation carriers we used the frequency of a trait in Congenita Project to collect clinical and genetic the KRT6A group as a reference when calculating data on patients with PC worldwide. The registry odds ratios (OR) for the same trait to occur in the was approved by the Western Institutional Review other keratin groups. To increase the power of Board (study #20040468). All patients gave written analysis, all family members having a PC phenotype informed consent and the study was conducted were included in the test, with intrafamilial correla- according to the Declaration of Helsinki tion adjusted. Software was used to perform the Principles. Participant enrollment began in May comparisons (STATA v9.2, StataCorp, College 2004. Station, TX). JAM ACAD DERMATOL Eliason et al 3

RESULTS Table I. International Pachyonychia Congenita An international case series of patients with Research Registry demographics mutation-veriﬁed PC No. of individuals At the time the data were collected, 254 individ- Mode of inheritance uals had completed the necessary steps for inclusion Familial 157 in the IPCRR (Table I). Additional demographics (eg, Spontaneous 97 country of residence) can be found at www. Gender pachyonychia.org. In addition to the 254 patients Male 121 harboring keratin mutations in one of the 4 ‘‘classic’’ Female 133 PC genes (KRT6A, KRT6B, KRT16, and KRT17)34 Location of residence other individuals were found to have no detectable United States 125 mutation, or mutations in other genes including Outside United States 129 connexin-30, KRT6C,23 and desmoglein-1. In this Age, y article we will focus on the results of those with Median 33 \ mutations in the 4 classic PC keratin genes. Range 1-86 We present the most commonly reported clinical ﬁndings according to mutation status in Table II.

manifestations of keratoderma, with calluses being Major phenotypic features of PC the most commonly reported (210/223, 94%) (data Three clinical features that were reported in more not shown). than 90% of patients across all mutation subtypes To better understand the persistence of the plan- were thickened toenails, plantar keratoderma, and tar lesions, patients were asked to approximate how plantar pain (Table II). long the lesions lasted: 228 of 254 (95%) indicated Thickened toenails. Thickened toenails (Fig 1) that their feet were always affected and that the were the most frequently reported clinical finding in lesions never completely resolved. Among 241 pa- the IPCRR with 249 of 254 (98%) patients reporting tients who reported the age of onset of their plantar this phenotype (Table II). The average number of keratoderma, the average age was 4.2 years with a toenails affected was 8.8 (range 0-10, mode = 10). We range of 0 (birth) to 30 years, and a median and mode performed logistic regression analysis to compare of 3 and 2 years, respectively. Of patients, 20% were the relative likelihood of having all 10 toenails affected by 1 year of age and 66% were affected by 4 affected between the different PC keratin mutation years of age (data not shown). carriers. Using the prevalence of 10 affected nails in We found that the KRT16 mutation carriers devel- participants with mutations in KRT6B, KRT16, and oped plantar keratoderma at a similar age to KRT6A KRT17 as the reference, we found that KRT6A carriers whereas KRT6B and KRT17 carriers were mutation carriers were 11.1 times as likely (P \ significantly more likely to report later onset. .001) to have all 10 toenails affected. The average age Plantar pain. Plantar pain was reported by 225 of onset of toenail dystrophy across all keratin of 254 (89%) surveyed patients. The prevalence of mutation types was 2.8 years with a median of 0.08 pain was high across keratin subgroups (Table II). years (1 month). KRT6A mutation carriers had the The age of the patient was found to have a dramatic earliest average onset at 0.35 years (about 4 months). effect on the reporting of plantar pain: only 3 patients The average age of onset for patients with a KRT6B, older than 10 years did not report plantar pain (data KRT16, and KRT17 mutations was 9.5, 6.8, and 0.9 not shown). Plantar pain was the most important years, respectively. feature of PC affecting quality of life (see below). Plantar keratoderma. Plantar keratoderma (Fig 2) was the next most commonly reported Other common clinical ﬁndings in PC finding, present in 241 of 254 (95%) patients (Table Fingernail involvement. As shown in Fig 3, II). The registry includes individuals of ages younger thickened fingernails were reported in 220 of 254 than 1 year to older than 86 years (Table I). Of the 13 (87%) patients (Table II). The prevalence was lower individuals who were not reported as having plantar in the other mutation carriers, with 9 of 20 (45%) keratoderma, 9 were younger than 1 year, and the KRT6B and 56 of 76 (74%) KRT16 mutation carriers oldest was 3 years of age (data not shown). Plantar reporting at least one affected nail (Table II). keratoderma can variably manifest as calluses, blis- Interestingly, when the number of fingernails af- tering, fissures, thickened skin, and open sores. Most fected was evaluated based on mutation type, pa- patients experienced more than one of these tients with KRT6B mutations appeared to have far 4 Eliason et al JAM ACAD DERMATOL

Table II. Prevalence of selected clinical features in pachyonychia congenita

Keratin gene affected Total KRT6A KRT16 KRT6B KRT17 Phenotype N = 254 N = 115 N=76 N=20 N=43 Thick toenails 249/254 (98%) 115/115 (100%) 71/76 (93%) 20/20 (100%) 43/43 (100%) Plantar keratoderma 241/254 (95%) 105/115 (91%) 76/76 (100%) 20/20 (100%) 40/43 (93%) Plantar pain 225/254 (89%) 101/115 (88%) 72/76 (95%) 20/20 (100%) 32/43 (74%) Thick fingernails 220/254 (87%) 115/115 (100%) 56/76 (74%) 9/20 (45%) 40/43 (93%) Oral leukokeratosis 177/254 (70%) 111/115 (97%) 45/76 (59%) 7/20 (35%) 14/43 (33%) Palmar keratoderma 149/254 (59%) 60/115 (52%) 63/76 (83%) 6/20 (30%) 20/43 (30%) Follicular hyperkeratosis 140/254 (55%) 88/115 (77%) 10/76 (13%) 7/20 (35%) 35/43 (81%) Hyperhidrosis 124/254 (49%) 60/115 (52%) 33/76 (43%) 10/20 (50%) 21/43 (49%) Cysts 104/254 (41%) 47/115 (41%) 8/76 (11%) 11/20 (55%) 38/43 (88%) Hoarseness 71/254 (28%) 49/115 (45%) 15/76 (20%) 7/20 (35%) 10/43 (23%) Natal teeth 39/254 (15%) 3/115 (3%) 0/76 (0%) 0/20 (0%) 36/43 (84%)

Fig 1. Clinical phenotype in genetically confirmed pachyonychia congenita. Toenail involvement usually demonstrates significant subungual hyperkeratosis and sometimes presents with premature termination of nail. fewer nails affected on average compared with those much higher likelihood of reporting cysts (OR of 23.2 with other keratin gene mutations. Comparison of [P = .003]) compared with KRT6A mutation carriers. the OR to develop fingernail involvement later than Natal teeth. The phenomenon of erupted teeth KRT6A mutation carriers revealed a statistically sig- present at birth known as ‘‘natal teeth’’ has been nificantly elevated OR for KRT6B and KRT16 but not reported in patients with PC. Of the 39 patients who for KRT17 carriers. reported the presence of teeth at birth, 36 were Mucosal involvement. Oral leukokeratosis was KRT17 mutation carriers. No KRT16 or KRT6B mu- reported in 177 of 254 individuals (70%) with the tation carriers reported natal teeth and only 3 of 115 breakdown by mutation noted in Table II. Of the 177 (3%) KRT6A carriers were affected (Table II). individuals with self-identified oral leukokeratosis, 125 reported an average age of onset of 5.1 years PC reduces the quality of life with a median and mode of 0 years. Onset at birth In Table III the severity and frequency of common was reported by 67 of 125 (54%), whereas 84 of 115 PC symptoms are detailed. (73%) reported onset affected by 1 year of age and Of the 240 patients who reported plantar pain, 214 119 of 125 (95%) by age 20 years (data not shown). indicated its frequency. A total of 138 (64%) indicated KRT6A and KRT17 carriers had a significantly in- that their quality of life was affected at least weekly creased OR of earlier onset of oral leukokeratosis by plantar pain. In all, 41 (20%) of respondents were compared with KRT6B and KRT16 carriers. affected ‘‘every month or two’’; 26 (12%) reported Cysts. Pilosebaceous cysts and steatocysts have being affected ‘‘seldom’’ (defined as ‘‘once a year or been reported in conjunction with PC types 1 and 2. less’’); and 9 (4%) indicated that they were never Overall, 104 of 254 (41%) patients reported cysts of affected. Describing the pain, 99 of 240 (41%) any type (Table II). KRT17 mutation carriers had a reported it was ‘‘very painful, but do not use JAM ACAD DERMATOL Eliason et al 5

Fig 2. Clinical phenotype in genetically conﬁrmed pachyonychia congenita. Plantar hyperkeratosis often follows pressure distribution, but can involve entirety of plantar surface. Environmental factors play signiﬁcant role in development and persistence of plantar keratoderma.

Fig 3. Clinical phenotype in genetically confirmed pachyonychia congenita. Fingernail findings include dramatic elevation of fingernails because of subungual debris or premature termination of nail plate. medication’’; 62 of 240 (26%) answered that they Because of difficulties inherent in the study of a ‘‘often require medication for the pain’’; and 79 of rare genodermatosis there are several limitations 240 (33%) reported that plantar pain was ‘‘somewhat with our data. The patients were ascertained either to not painful.’’ by self- or physician-referral, not from a population- based assessment. Because most self-referrals were DISCUSSION from individuals who became aware of PC through This cohort includes patients with the 4 most the Internet, our cohort probably represents a more common PC keratin mutations in sufficient numbers affluent, better-educated population with access to to draw conclusions regarding the prevalence and health care. Because of the geographically dispersed penetrance of the most common PC clinical findings. nature of the cohort, our data were mostly gathered To out knowledge, it includes the most diverse through telephone interviews and evaluation of collection of patients of any cohort thus far evaluated photographs, rather than by direct examination, and reduces the impact of genetic background or which makes the objective quantification of clinical founder bias among the clinical features reported for severity more difficult. The quality-of-life measures different mutations. and reporting of pain, although key to 6 Eliason et al JAM ACAD DERMATOL

Table III. Impact of specific features of pachyonychia congenita on quality of life

Impact on quality of life Plantar keratoderma Thickened toenails Cysts Thickened fingernails Oral leukokeratosis No impact 19/216 (9%) 71/221 (32%) 63/154 (41%) 67/199 (34%) 40/64 (63%) Sometimes a problem 56/216 (26%) 113/221 (51%) 65/154 (42%) 109/199 (55%) 23/64 (35%) Always a problem, but 133/216 (62%) 36/221 (16%) 23/154 (15%) 22/199 (11%) 1/64 (2%) able to function Unable to function 8/216 (4%) 1/221 (0.5%) 3/154 (2%) 1/199 (0.5%) 0/64 (0%)

understanding the most important issues that affect to find and maintain work. The pain reported by patients with PC, were not performed using a vali- patients with PC is often out of proportion to the dated metric. extent or duration of callus, suggesting that the The most common clinical findings in our PC mechanism is not merely the result of pressure cohort were toenail thickening, plantar keratoderma, from callus formation. Clinicians will better meet and plantar pain. Only one patient, who was youn- the needs of patients with PC by inquiring about the ger than 1 year, lacked all 3 findings. The impact of extent of pain and helping them manage their age on their prevalence is significant as children calluses to facilitate pain reduction. seldom develop plantar keratoderma or plantar pain Historically, PC has been subdivided into two before they start to walk. If children 3 years of age major phenotypic variants, PC-1 (Jadassohn- and younger were excluded from the analysis, 216 of Lewandowski) and PC-2 (JacksoneLawler). The 230 (94%) individuals would meet all 3 criteria and PC-1/PC-2 classification was designed to improve 227 of 230 (99%) of our participants would have met the ability to predict phenotypic prognosis without two criteria. For the clinician, these findings suggest genetic testing. The PC-1/PC-2 classification assumes that the presence of toenail dystrophy with plantar that because certain keratin proteins predictably keratoderma and plantar pain in children older than dimerize that a mutation in either protein gene will 3 years is a much more sensitive means of clinically result in a similar clinical phenotype. Hence, a diagnosing PC than 20-nail dystrophy, which is often mutation in either of the PC-1 keratin proteins thought to be requisite to the diagnosis. (K6a/K16) should cause similar features, whereas a Additional diagnostic findings associated with PC mutation in K6b/K17 (PC-2 proteins) will present included fingernail dystrophy, follicular hyperkera- with a different, predictable phenotype. tosis, leukokeratosis, cysts, and natal teeth. Less Instead, our data demonstrated that clinical phe- common findings included ear pain, hoarseness, notypes overlapped substantially across genotypic and hyperhidrosis. In the literature there are reports categories and could not be used to predict genotype of corneal findings, deafness, skeletal abnormalities, reliably. Specifically, we found a significant overlap and mental retardation associated with PC. We found of oral leukokeratosis, cysts, and natal teeth that no support for these findings among our cohort. A purportedly distinguish PC-1 and PC-2. Overall, our more detailed discussion of features spuriously as- data demonstrate that the PC-1/PC-2 nomenclature sociated with PC will be reported elsewhere. does not accurately reflect the molecular pathogen- Although a detailed discussion of the clinical esis of PC and does not represent a rational or findings linked to a specific mutation is beyond the clinically useful classification at this time. scope of this article, we found that the KRT16 We recommend the elimination of the terms mutations, p.Asn125Ser and p.Arg127Cys, were ‘‘PC-1’’ and ‘‘PC-2’’ and propose their replacement strongly associated with lack of fingernail involve- with notation of the specific keratin defect. In this ment whereas KRT16 mutation carriers with the classification scheme a diagnosis of PC-6a, PC-6b, p.Leu132Pro mutation frequently presented with PC-16, and PC-17 would correspond to mutations in 10-fingernail dystrophy. Our group has recently the KRT6A, KRT6B, KRT16, and KRT17 genes, published a large study reviewing new and previ- respectively. A designation of PC-U (unknown) ously known mutations in patients with PC.24 As may be applied when the classic clinical findings of more data are gathered we hope to be able to PC are found in the absence of a known PC keratin provide phenotypic prognosis based on the specific gene mutation. Classification based on keratin mu- mutation identified. tation subtype will allow clinicians to provide more Plantar pain has the most profound effect on accurate prognoses for patients with PC. We recom- quality of life for most patients with PC as it can limit mend genetic testing for individuals with the triad of mobility and social interaction along with the ability toenail dystrophy, plantar keratoderma, and plantar JAM ACAD DERMATOL Eliason et al 7

pain as they have a high likelihood of carrying a PC 12. Schonfeld PH. The pachyonychia congenita syndrome. Acta mutation. Genetic testing is provided free of charge Derm Venereol 1980;60:45-9. to all patients who enroll in the PC registry through 13. Franzot J, Kansky A, Kavcic S. Pachyonychia congenita (Jadassohn-Lewandowsky syndrome): a review of 14 cases in the nonprofit patient advocacy group, Pachyonychia Slovenia. Dermatologica 1981;162:462-72. Congenita Project (www.pachyonychia.org). 14. Stieglitz JB, Centerwall WR. Pachyonychia congenita (Jadas- We look forward to the development of speciﬁc sohn-Lewandowsky syndrome): a seventeen-member, genetic techniques to minimize or eliminate the four-generation pedigree with unusual respiratory and dental clinical expression of this rare keratin disorder.25-27 involvement. Am J Med Genet 1983;14:21-8. 15. Sivasundram A, Rajagopalan K, Sarojini T. Pachyonychia congenita. Int J Dermatol 1985;24:179-80. REFERENCES 16. Feinstein A, Friedman J, Schewach-Millet M. Pachyonychia 1. Smith FJD, Kaspar RL, Schwartz ME, McLean WHI, Leachman congenita. J Am Acad Dermatol 1988;19:705-11. SA. Pachyonychia congenita. GeneReviews 2006. Available 17. Su WP, Chun SI, Hammond DE, Gordon H. Pachyonychia from: URL: http://www.genetests.org/profiles/pc. Accessed congenita: a clinical study of 12 cases and review of the October 15, 2010. literature. Pediatr Dermatol 1990;7:33-8. 2. Leachman SA, Kaspar RL, Fleckman P, Florell SR, Smith FJ, 18. Paller AS, Moore JA, Scher R. Pachyonychia congenita tarda: a McLean WH, et al. Clinical and pathological features of pachy- late-onset form of pachyonychia congenita. Arch Dermatol onychia congenita. J Investig Dermatol Symp Proc 2005;10:3-17. 1991;127:701-3. 3. McLean WH, Rugg EL, Lunny DP, Morley SM, Lane EB, 19. Dahl PR, Daoud MS, Su WP. Jadassohn-Lewandowski syndrome Swensson O, et al. Keratin 16 and keratin 17 mutations cause (pachyonychia congenita). Semin Dermatol 1995;14:129-34. pachyonychia congenita. Nat Genet 1995;9:273-8. 20. Munro CS, Carter S, Bryce S, Hall M, Rees JL, Kunkeler L, et al. A 4. Bowden PE, Haley JL, Kansky A, Rothnagel JA, Jones DO, gene for pachyonychia congenita is closely linked to the Turner RJ. Mutation of a type II keratin gene (K6a) in keratin gene cluster on 17q12-q21. J Med Genet 1994;31: pachyonychia congenita. Nat Genet 1995;10:363-5. 675-8. 5. Smith FJ, Jonkman MF, van Goor H, Coleman CM, Covello SP, 21. van Steensel MA, Jonkman MF, van Geel M, Steijlen PM, Uitto J, et al. A mutation in human keratin K6b produces a McLean WH, Smith FJ. Clouston syndrome can mimic pach- phenocopy of the K17 disorder pachyonychia congenita type yonychia congenita. J Invest Dermatol 2003;121:1035-8. 2. Hum Mol Genet 1998;7:1143-8. 22. Munro CS. Pachyonychia congenita: mutations and clinical 6. Kaspar RL. Challenges in developing therapies for rare diseases presentations. Br J Dermatol 2001;144:929-30. including pachyonychia congenita. J Invest Dermatol Symp 23. Wilson NJ, Messenger AG, Leachman SA, O’Toole EA, Lane EB, Proc 2005;10:62-6. McLean WH, et al. Keratin K6c mutations cause focal palmo- 7. Muller C. Zur Kasuistic der kongenitalen onchogryphosis [On plantar keratoderma. J Invest Dermatol 2010;130:425-9. the causes of congenital onychogryposis]. Muenchener Med- 24. Wilson NJ, Leachman SA, Hansen CD, McMullan AC, Milstone izinische Wochenschrift 1904;49:2180-2. LM, Schwartz ME, et al. A large mutational study in pachyo- 8. Wilson AG, Cantar MB. Three cases of hereditary hyperkerato- nychia congenita. J Invest Dermatol 2011;131:1018-24. sis of the nail-bed. Br J Dermatol 1904;17:13-4. 25. Leachman SA, Hickerson RP, Hull PR, Smith FJ, Milstone LM, 9. Jadassohn J, Lewandowski P. Pachyonychia congenita: kera- Lane EB, et al. Therapeutic siRNAs for dominant genetic skin tosis disseminata circumscripts (follicularis). Tylomata. Leuko- disorders including pachyonychia congenita. J Dermatol Sci keratosis linguae. In: Neisser A, Jacobi E, editors. Ikonographia 2008;51:151-7. dermatologica. Berlin: Urban and Schwarzenberg; 1906. pp. 26. Leachman SA, Hickerson RP, Schwartz ME, Bullough EE, 29-31. Hutcherson SL, Boucher KM, et al. First-in-human 10. Kumer L, Loos HO. Ueber pachyonychia congenita (typus mutation-targeted siRNA phase Ib trial of an inherited skin Riehl) [On pachyonychia congenita (Riehl type)]. Wien Klin disorder. Mol Ther 2010;18:442-6. Wochenschr 1935;48:174-8. 27. Hickerson RP, Smith FJ, Reeves RE, Contag CH, Leake D, 11. Moldenhauer E, Ernst K. Das Jadassohn-Lewandowsky- Leachman SA, et al. Single-nucleotide-specific siRNA targeting syndrom [The Jadassohn-Lewandowsky syndrome]. Hautarzt in a dominant-negative skin model. J Invest Dermatol 2008; 1968;19:441-7. 128:594-605.

DOI: 10.1111/jdv.12098 JEADV

ORIGINAL ARTICLE Best treatment practices for pachyonychia congenita

I. Goldberg,1,* D. Fruchter,1 A. Meilick,1 M.E. Schwartz,2 E. Sprecher1 1Department of Dermatology, Tel-Aviv Sourasky Medical Center, Tel-Aviv, Israel 2Pachyonychia Congenita Project, Salt Lake City, UT, USA *Correspondence: I. Goldberg. E-mail: [email protected]

Abstract Background Numerous therapeutic modalities have been proposed to treat the manifestations of pachyonychia congenita (PC). While research hopes lie with molecular therapies, patients are in need of answers regarding the efficacy of conventional treatments. Aim of the study To determine patients’ experience and preferences regarding conventional treatments for PC. Methods The study population included 120 PC patients from 20 countries. The study was based on a patient survey developed by physicians and researchers from the International Pachyonychia Congenita Consortium and conducted via the internet. Using an effectiveness scale of 1 to 5, the patients were asked to grade treatments for different manifestations, including keratoderma, cysts, follicular hyperkeratosis, fingernail and toenail involvement. Results Patients reported surgical treatments being most effective for cysts and mechanical treatments the most effective conventional therapeutic approach for all other investigated manifestations. The other conventional medical treatments were found to be non-effective to only slightly effective. Among patients with keratoderma, older people were more likely to report beneficial effect from mechanical treatments (P = 0.04), topical retinoids (P = 0.04) and topical steroids (P = 0.02). Likewise, females were more inclined to report filing and grinding beneficial than males (P = 0.02). Finally, carriers of KRT16 and KRT6a were more likely to benefit from keratolytics than carriers of mutations in KRT17 (P = 0.04). Conclusions None of the currently available therapeutic options for PC are ideal, although they provide some relief, with mechanical/surgical options being preferred over medical therapies. These results emphasize the need for more efficient and targeted therapies. Received: 5 July 2012; Accepted: 18 December 2012

Conﬂict of interest None declared.

Funding sources Funded by the Pachyonychia Congenita Project.

Introduction hyperkeratosis, palmar keratoderma, epidermal cysts, oral leu- PC is a very rare keratinizing disorder estimated to affect kokeratosis, and occasionally hyperhidrosis, hoarseness and between 5 and 10 thousand people worldwide.1 This disorder natal teeth.1,2 is transmitted as an autosomal dominant trait, and is caused Clinical classification of PC variants was first suggested by by mutations in one of five keratin genes: KRT6A, KRT6B, Kumer in 1935.8 PC was eventually divided into two clinical sub- KRT6C, KRT16, KRT17, which encode keratins K6a, K6b, types: the Jadassohn-Lewandowski PC (type-1 PC) and the K6c, K16 and K17 respectively.1,2 Most of the keratin muta- Jackson-Lawler PC (type-2 PC).1,2 tions which cause PC are heterozygous missense mutations or This clinical classification was intended to assist estimation of small insertions/deletions which result in fragility of the epi- the prognosis in the absence of genetic testing.1,2 After the dis- thelial cell cytoskeleton, leading to cell cytolysis and tissue covery of the underlying cause of PC, genotype-phenotype anal- – blistering or hyperkeratosis.2 4 ysis suggested initially that mutations in KRT6a/KRT16 and PC cardinal features were first reported by Muller5 and KRT6b/KRT17 were associated with type 1 and type 2 PC Wilson6 in 1904, and by Jadassohn and Lewandowski in respectively.1 More recently, large-scale genetic analysis raised 1906,7 and include painful and debilitating plantar keratoder- doubts regarding the clinical relevance of these correlations,1,2 ma, hypertrophic toenail and fingernail dystrophy, follicular leading to the establishment of a novel classification for PC

based solely on molecular analysis resulting in PC-K6a, PC-K6b, removal of the lesions, surgical removal of the nail, incision PC-K6c, PC-K16 and PC-K17 as distinct types.1,2 and drainage of cysts, soaking of the nails to soften them Many therapeutic modalities have been proposed to treat before treatments, orthotics, custom made orthotics, topical the various clinical manifestations in PC, including retinoids,9 and oral retinoids, pain medications, botulinum toxin, mois- surgical and mechanical procedures, orthotics, keratolytics, turizers, vaseline, keratolytic treatments, antibiotic ointments, pain medications10 and botulinum toxin.11 Recently, more antifungal ointments, topical and oral steroids, salicylic acid, targeted therapeutic strategies (including small interfering treatments of the nails by medical professional and treatments – RNAs,12 14 rapamycin15 and simvastatin16) have been the of the nails in a nail salon. focus of much attention. Unfortunately, as most of these advanced approaches still cannot be offered on a routine Statistical analysis basis to patients because of expense and limited availability, All outcome variables are ordinal variables ranging from 1 to 5 – patients are currently forced to rely upon the use of the the higher the score, the higher the treatment effect. Univariate available conventional strategies. analysis was used to determine the relationships between each In this study, we used a patient survey approach to derive explanatory variable and the treatment outcome variables. The effectiveness data in PC. This methodology has been widely used explanatory variables are the following: age, gender, quality of over the past years to delineate guidelines for the treatment of life score (QOL) and gene (categorical variable). disorders for which no controlled data are available.9 Pearson correlations were calculated between all continuous explanatory variables and the outcome variables. Wilcoxon Two- Methods Sample or Kruskal–Wallis tests were used to compare between categorical explanatory variables and the outcome variables. Patients A P-value of 0.05 was considered significant. Statistical analy- The study population included pachyonychia congenita patients sis was performed by SAS for windows version 9.2 (SAS Insti- from 20 different countries who were enrolled in the Interna- tute, Cary, NC, USA). tional Pachyonychia Congenita Research Registry (IPCRR) and recruited through Pachyonychia Congenita Project, a non-profit Results patient advocacy group established in 2004.1,2 All patients were The study included 120 PC patients, 67 females and 53 males, of diagnosed using a detailed clinical questionnaire and genetic all ages. The youngest patient was a 1-year-old baby and the old- testing results for a mutation in one of the PC-associated genes. est an 81-year-old patient. The average age was 38.5, and the The study was conducted according to the principles of the dec- median age was 39 years. laration of Helsinki and all patients gave their written informed consent. Keratoderma Of the 120 patients who took part in the study, 113 patients Data collection reported having keratoderma. Forty-eight patients had clinical In addition to the extensive physician-validated data in the manifestations involving both palms and soles; 65 patients had IPCRR, each patient completed an addendum survey via the in- only sole involvement. ternet providing information on treatments used for five catego- Most conventional treatments were attributed mean scores of ries of clinical manifestations of PC: keratoderma, cysts, 2 to 3 (a little effective to somewhat effective) (Fig. 1). Patients follicular hyperkeratosis, fingernail and toenail involvement. In reported mechanical treatments (such as filing, grinding, cutting, addition, all patients provided information on demographics, clipping), as the most effective conventional treatments corre- genetic status, the effect of the disease on their quality of life, the sponding to a mean score of 3.8. Pain control medications and clinical manifestations of PC and the degree of effectiveness of orthotics ranked second in effectiveness. the different treatments. Other medications such as antibiotic ointments, urea cream, The patients were asked to grade each treatment they had oral and topical retinoids, oral and topical steroids, botulinum used according to a treatment effectiveness scale of 1 to 5: toxin injections, salicylic acid or antifungal treatments were 1- not effective at all reported as poorly effective. 2- a little effective 3- somewhat effective Cysts 4- effective Of the 120 patients who took part in the study, 49 patients 5- very effective. reported having cysts (Fig. 2). Surgical removal treatment was The patients were asked to grade different available treat- found to be effective to very effective, reaching a mean score of ments including mechanical treatments (such as filing, grind- 4.32. The next most effective treatment reported was incision ing, cutting, clipping or plucking the lesions), surgical and drainage with a mean score of 4. Pain medications were

Mean scores of treatment effect in keratoderma 5 4.5 3.8 4 3.5 3.0 3.0 2.7 3 2.6 2.5 2.5 2.5 2.2 2.1 2.1 2.1 1.9 1.9 2 1.7 1.3 1.3 1.5 1.1 Effectiveness score 1 0.5 0

Figure 1 Mean scores of effectiveness of the different treatments used for keratoderma (reported by 113 patients). Treatment effectiveness is represented by a colour scale: red - not effective treatment (scores between 1 and 2), orange - mildly effective treatment (scores between 2 and 3), green - somewhat effective treatment (scores between 3 and 4), blue - effective treatment (scores between 4 and 5).

Mean scores of treatment effect in cysts 5 4.32 4.5 4 4 3.5 3 2.38 2.2 2.5 2 1.6 1.5 1.4 1.3 1.25 1.5

Effectiveness score 1 0.5 0

Figure 2 Mean scores of effectiveness of the different treatments for cysts (reported by 49 patients). Treatment effectiveness colour scale is described in ﬁgure legend 1.

found to be of little effect. Other treatments, such as antibiotic Toenail involvement ointments, intralesional and oral steroids and retinoids, were Toenail involvement is very common in PC. Of the 120 patients reported as not effective. in the study, 108 patients reported toenail involvement. The results for toenail involvement (Fig. 4) demonstrated that the Follicular hyperkeratosis most effective approach was mechanical treatment such as ﬁling, Of the 120 patients who took part in the study, 63 patients grinding, cutting or clipping the toenails. Mechanical treatment reported being affected with follicular hyperkeratosis. The received a mean score of 4. The patients reported that the second results (Fig. 3) showed that the available conventional treat- most effective treatment, with a mean score of 3.6, was soaking ments for follicular hyperkeratosis were slightly effective or the nails to soften them before treatment. Of note, surgical avul- not effective, with clipping or plucking the plugs leading with sion of the nail was found to be ineffective, as were non-medical a mean score of 2.9. treatments provided in nail salons.

5 Mean scores of treatment effect in follicular hyperkeratosis

4.5

3.5 2.9 3 2.47

2.5 2.19 2.13

Effectiveness score 1.67 2 1.56 1.33 1.5

0.5

Figure 3 Mean scores of effectiveness of the different treatments used for follicular hyperkeratosis (reported by 63 patients). Treatment effectiveness colour scale is described in ﬁgure legend 1.

5 Mean scores of treatment effect in toenail involvement

4.5 4.06

4 3.67

3.5 3.14

2.34 2.5 2.25 2.21 2.07 2 2 1.66 1.61 1.6 Effectiveness score 1.47 1.5

0.5

Figure 4 Mean scores of effectiveness of the different treatments used for toenail involvement (reported by 108 patients). Treatment effectiveness colour scale is described in ﬁgure legend 1.

Fingernail involvement second most effective treatment was soaking and softening Fingernail involvement was reported by 94 of the 120 patients in the nails. The other treatments ascertained were found to be this study. Results pertaining to fingernail involvement resem- not effective. bled those obtained for toenail involvement (Fig. 5). The most effective treatment was mechanical treatment Univariate analysis such as grinding, filing, clipping or cutting the fingernails. As multivariate analysis was found to be very unstable due to the Mechanical treatments received a mean score of 4 and the wide range of available data, we used univariate analysis to

Mean scores of treatment effect in fingernail involvement 5

4.5 4.01 4

3.5 3.25

eroc 3

ssse 3 2.5 nevitceffE 2.5 2.32 2.17 2.05 2 1.82 1.6 1.4 1.38 1.5 1.08 1

0.5

Figure 5 Mean scores of effectiveness of the different treatments used for ﬁngernail involvement (reported by 94 patients). Treatment effectiveness colour scale is described in ﬁgure legend 1.

further analyse the data and to determine the relationships ablation is performed. Follicular hyperkeratosis can be treated between each explanatory variable and the treatment outcome by oral and topical retinoids, keratolytic agents and alpha- variables. This analysis revealed three facts of clinical relevance hydroxy acid preparations. Cysts may be treated by incision, for patients with keratoderma: excision, drainage or by intralesional injection of steroids. In 1 As patients grew older, they were more likely to report benefi- case of infection, oral antibiotics may be indicated.10 cial effect from mechanical interventions (such as filing, Currently, retinoids are considered efficient drugs to treat grinding) (P = 0.04), topical retinoids (P = 0.04) and topical hyperkeratotic disorders including PC.9,16 They act via retinoic steroids (P = 0.02). acid response elements (RAREs) which are present in the kera- 2 Females were more inclined than males (P = 0.02) to tin’s gene promoters, and inhibit gene expression.16 Retinoids, describe mechanical interventions such as filing and grinding although reducing hyperkeratosis, may also cause thinning of as beneficial. the epidermis and blistering, leading to pain and possible infec- 3 Finally, carriers of mutations in KRT16 and KRT6a were tious complications.10,16,18 Contradictory data regarding the effi- more likely to benefit from keratolytics than carriers of muta- cacy of retinoids in PC have been published. Some case reports tions in KRT17 (P = 0.04). demonstrated an improvement of calluses with retinoid treat- – ment.19 22 Other studies described patients with improvement Discussion of hyperkeratosis with retinoid treatment, but no change in Treating PC is challenging. PC is clinically multifaceted and the pachyonychia.21,22 More recently, Gruber et al.9 analysed data conventional treatments are directed at the different manifesta- collected in 30 PC patients who received systemic retinoid treat- tions of the disorder. Currently, there are no specific treatments ment. They found that 50% and 14% of their patients reported for PC.16,17 Each patient presents a unique constellation of con- improvement in palmoplantar hyperkeratosis and pachyonychia ditions and a treatment plan must be individually tailored.17 respectively. The mean satisfaction score from the treatment was Unfortunately, despite use of numerous approaches to relieve found to be 4.5 on a scale of 1–10. All patients reported suffering PC-associated symptoms, little is currently known regarding from adverse effects and 83% stopped using the drug.9 In con- their relative efficacy. trast with these mixed results, others have reported no improve- Conventional treatment for nail disease in PC includes ment with retinoids in PC23,24 and, in another series,25 four mechanical or surgical procedures, such as grooming or surgical patients with palmoplantar keratoderma who received oral reti- removal of nails. The nails tend to re-grow unless complete noids reported improvement in the appearance of their skin, but

had to stop the treatment because of pain that restricted hand tance. For example, a mutation-specific siRNA was found and foot function. recently to lead to callus recession and pain control in a PC – Levels of evidence vary according to the methodology used in patient.12 14,16 However, these injections were found to be tre- clinical studies, which in turn is often a function of patient pop- mendously painful underscoring the need for new approaches ulation size. Randomized control studies are of course prefera- for more efficient and practical ways of nucleic acid delivery to ble; however, in diseases as infrequent as PC, such studies are the skin.27 Other treatments under investigation for PC patients not always possible.9,18 Alternative methodologies, also adapted are rapamycin and simvastatin, but these therapies, although to rare conditions in which controlled studies cannot be readily providing hope for patients, are not yet applied routinely.10,15,16 performed, rely upon the quantification of patient values and In conclusion, although none of the currently available thera- expectations. peutic approaches seem to be ideal, they do provide some relief Our study was based on a survey conducted among the largest to our patients. Mechanical/surgical options are preferred over group of PC patients ascertained to date for treatment efficacy. medical therapies, such as retinoids, antibiotics or antifungal As all patients had been diagnosed with PC on the basis of both agents. These results emphasize the need for more efficient and a careful physical examination and a full molecular analysis, we targeted therapies. believe that the data collected faithfully reflect PC patients’ appreciation of conventional therapeutic modalities. Acknowledgements We found that the majority of conventional treatments were We thank patients and their families in the International Pachy- only marginally effective for keratoderma, with mechanical treat- onychia Congenita Research Registry (IPCRR) for their support ments being the most effective. and members of the International Pachyonychia Congenita Con- Palmoplantar keratoderma is a very common manifestation soritum (IPCC) for development of the treatment survey. of PC, usually presenting as a child starts walking and bearing weight during the first few years of life.10 In a study by Eliason, References et al.1 plantar keratoderma was present in 241 of 254 (95%) of 1 Eliason MJ, Leachman SA, Feng BJ, Schwartz ME, Hansen CD. A review patients. Of the 13 patients reported as not having plantar kera- of the clinical phenotype of 254 patients with genetically confirmed pachyonychia congenita. J Am Acad Dermatol 2012; 67: 680–686. toderma, the oldest was 3-years old and nine were younger than 2 McLean WH, Hansen CD, Eliason MJ, Smith FJ. The phenotypic and 1 year. Among 241 patients who reported having plantar kerato- molecular genetic features of pachyonychia congenita. J Invest Dermatol derma, the age of onset ranged from birth to 30 years and the 2011; 131: 1015–1017. 3 McLean WH, Smith FJ, Cassidy AJ. Insights into genotype-phenotype average age was 4.2 years. In our study, most of the patients who correlation in pachyonychia congenita from the human intermediate did not report plantar keratoderma were younger than 5 years filament mutation database. J Investig Dermatol Symp Proc 2005; 10: of age. 31–36. Patients reported surgical treatments being most effective for 4 McLean WH, Moore CB. Keratin disorders: from gene to therapy. Hum Mol Genet 2011; 20: R189–R197. cysts. Mechanical treatments were found to be the most effective 5 Muller C. Zur Kasuistic der kongenitalen onchogryphosis [on the causes conventional therapeutic approach for follicular hyperkeratosis, of congenital onychogryposis]. Muenchener Medizinische Wochenschrift as well as for toenail and fingernail involvement. The other con- 1904; 49: 2180–2182. ventional medical treatments were found to be non-effective to 6 Wilson AG, Cantar MB. Three cases of hereditary hyperkeratosis of the nail-bed. Br J Dermatol 1904; 17:13–14. only slightly effective. 7 Jadassohn J, Lewandowski P. Pachyonychia congenita: keratosis dissemi- 26 In line with a previous study, keratolytics, which are widely nate circumscripts (follicularis). Tylomata. Leukokeratosis linguae. In used by the patients, were found to be of limited effectiveness Neisser A, Jacobi E, eds. Ikonographia Dermatologica. Urban and Sch- – for both palmoplantar keratoderma and nail problems. warzenberg, Berlin, 1906: 29 31. 8 Kumer L, Loos HO. Ueber pachyonychia congenita (typus Riehl)[on Response to treatment was highly individual, underscoring pachyonychia congenita (Riehl type)]. Wien Klin Wochenschr 1935; 48: the need for clinical or molecular predictors of response to ther- 174–178. apy. In this regard, the results of our univariate analysis, which 9 Gruber R, Edlinger M, Kaspar RL et al. An appraisal of oral retinoids in 66 remain to be independently confirmed, suggest a number of the treatment of pachyonychia congenita. J Am Acad Dermatol 2012; : e193–e199. such predictive parameters. For example, our finding that carri- 10 Smith FJD, Hansen CD, Hull PR et al. Pachyonychia congenita. In Pa- ers of mutations in KRT16 and KRT6a were more likely to bene- gon RA, Bird TD, Dolan CR, Stephens K, Adam MP, eds. GeneRe- fit from keratolytics than carriers of mutations in KRT17 may be viewsTM [Internet], University of Washington, Seattle, Seattle (WA), explained by the fact that the latter group has milder kerato- 2006. 1 11 Swartling C, Karlqvist M, Hymnelius K, Weis J, Vahlquist A. Botulinum derma than the former groups. This finding is consistent with toxin in the treatment of sweat-worsened foot problems in patients with phenotypic differences. epidermolysis bullosa simplex and pachyonychia congenita. Br J Dermatol Overall, current available therapeutic approaches in PC are of 2010; 163: 1072–1076. 12 Kaspar RL, Leachman SA, McLean WH, Schwartz ME. Toward a treat- borderline benefit. Therefore, the recent major advances in the ment for pachyonychia congenita: report on the 7th Annual International search for PC-specific therapeutic strategies are of great impor-

Pachyonychia Congenita Consortium meeting. J Invest Dermatol 2011; 20 Carabott F, Archer CB, Griffiths WA. Etretinate-responsive pachyonychia 131: 1011–1014. congenita. Br J Dermatol 1988; 119: 551–553. 13 Hickerson RP, Smith FJD, Reeves RE et al. Single-nucleotide-specific siR- 21 Lim TW, Paik JH, Kim NI. A case of pachyonychia congenita with NA targeting in a dominant-negative skin model. J Invest Dermatol 2008; oral leukoplakia and steatocystoma multiplex. J Dermatol 1999; 26: 128: 594–605. 677–681. 14 Leachman SA, Hickerson RP, Hull PR et al. Therapeutic siRNAs for dom- 22 Hoting E, Wassilew SW. [Systemic retinoid therapy with etretinate in inant genetic skin disorders including pachyonychia congenita. J Dermatol pachyonychia congenita]. Hautarzt 1985; 36: 526–528. [German]. Sci 2008; 51: 151–157. 23 Thomas DR, Jorizzo JL, Brysk MM, Tschen JA, Miller J, Tschen EH. 15 Hickerson RP, Leake D, Pho LN, Leachman SA, Kaspar RL. Rapamycin Pachyonychia congenita. Electron microscopic and epidermal glycopro- selectively inhibits expression of an inducible keratin (K6a) in human tein assessment before and during isotretinoin treatment. Arch Dermatol keratinocytes and improves symptoms in pachyonychia congenita 1984; 120: 1475–1479. patients. J Dermatol Sci 2009; 56:82–88. 24 Soyuer U, Candan MF. Failure of etretinate therapy in pachyonychia con- 16 Zhao Y, Gartner U, Smith FJ, McLean WH. Statins downregulate K6a genita. Br J Dermatol 1987; 117: 264. promoter activity: a possible therapeutic avenue for pachyonychia con- 25 Fritsch P, Honigsmann€ H, Jaschke E. Epidermolytic hereditary palmopl- genita. J Invest Dermatol 2011; 131: 1045–1052. antar keratoderma. Report of a family and treatment with an oral 17 Milstone LM, Fleckman P, Leachman SA et al. Treatment of pachyony- aromatic retinoid. Br J Dermatol 1978; 99: 561–568. chia congenita. J Investig Dermatol Symp Proc 2005; 10:18–20. 26 Su WP, Chun SI, Hammond DE, Gordon H. Pachyonychia congenita: a 18 Ormerod AD, Campalani E, Goodfield M. British Association of Derma- clinical study of 12 cases and review of the literature. Pediatr Dermatol tologists guidelines on the efficacy and use of acitretin in dermatology. 1990; 7:33–38. Br J Dermatol 2010; 162: 952–963. 27 Hickerson RP, Flores MA, Leake D et al. Use of self-delivery siRNAs to 19 Dupre A, Christol B, Bonafe JL, Touron P. [Pachyonychia congenita. inhibit gene expression in an organotypic pachyonychia congenita model. Three familial cases. Effects of the treatment by aromatic retinoid (RO J Invest Dermatol 2011; 131: 1037–1044. 10.9359)]. Ann Dermatol Venereol 1981; 108: 145–149. [French].

Dermatologic Therapy, Vol. 00, 2015, 00–00 VC 2015 Wiley Periodicals, Inc. DOI:10.1111/12297 DERMATOLOGIC THERAPY ISSN 1396-0296 THERAPEUTIC HOTLINE

Efficacy of botulinum toxin in pachyonychia congenita type 1: report of two new cases

Jessica Gonzalez-Ramos, Elena Sendagorta-Cudos, Guillermo Gonzalez-L opez, Ander Mayor-Ibarguren, Rosa Feltes-Ochoa & Pedro Herranz-Pinto Department of Dermatology, La Paz University Hospital, Madrid, Spain

ABSTRACT: Pachyonychia congenita (PC) is a rare genodermatosis caused by a mutation in keratin genes, which can lead to hypertrophic nail dystrophy and focal palmoplantar keratoderma (predominantly plantar), amongst other manifestations. Painful blisters and callosities, sometimes exacerbated by hyperhidrosis, are major issues that can have a significant impact on patient quality of life. Many alternative treatments for this condition have been applied with variable and partial clinical response, but a definitive cure for this disease has yet to be discovered. After obtaining informed consent, two patients with genetically confirmed PC type 1 were treated with plantar injections of botulinum toxin type A. Both patients showed a marked improvement in pain and blistering with an average response time of one week, a six-month mean duration of effectiveness, and a lack of any side effects or tachyphylaxis.

KEYWORDS: botulinum toxin, hyperhidrosis, pachyonychia congenita

Introduction painful focal areas of hyperkeratosis and blistering, particularly on the soles of the feet (plantar Pachyonychia congenital (PC) is a group of auto- keratoderma), and occasionally by oral leukoker- somal dominant congenital keratinopathies atosis, follicular hyperkeratosis, laryngeal caused by mutations in any of five genes: KRT6A, involvement, or other ectodermal defects (cysts KRT6B, KRT6C, KRT16, or KRT17. With a preva- or natal teeth). lence of 1 : 1,00,000 for this condition, there are Diagnosis of PC is determined through clinical estimated to be 5,000–10,000 cases worldwide. examination and confirmed by molecular genetic Historically PC has been broken down into testing, which is provided free to all patients two clinical subtypes: PC type 1 (Jadassohn- enrolled in the PC registry (PC Project) http:// Lewandowski) and PC type 2 (Jackson-Lawler), but www.pachyonychia.org (1). clinical phenotypes overlap between both types. Painful foot blistering and callosities are com- PC-1 is characterized by thickened toenails mon problems in patients with PC which are and fingernails in the first stages, extremely often exacerbated by hyperhidrosis and high ambient temperature. Hyperhidrosis can be Address correspondence and reprint requests to: Jessica blocked by plantar injections of botulinum toxin Gonzalez-Ramos, MD, Department of Dermatology, La Paz (BTX) (2). We report two patients who experi- University Hospital, Paseo La Castellana 261, CP 28046, enced remarkable improvement in terms of Madrid, Spain, or email: [email protected]. reduced plantar pain and foot blistering after BTX

1 Gonzalez-Ramos et al.

FIG. 1. Case 1 (A) Focal plantar keratoderma, blisters, and erosions on the left plantar surface before botulinum toxin injections. (B) Improve- ment after plantar injections of BTX with clearance of blisters and erosions. injections without loss of efficacy thus far. Anaes- palm. Postinjection clinical improvement began thesia was achieved in both cases via superficial in week one. Plantar pain and blistering were sig- sedation with spontaneous ventilation. nificantly mitigated and the patient was able to mobilize without her wheelchair (FIG. 1B). The effect of this treatment lasted approximately 6 Case reports months, so injections were subsequently administered biannually. In later sessions, the dosage was Case 1 reduced (75 U in each sole and 50 U in each The first patient, a 32-year-old woman, had been palm), concentrating the injections below blisters diagnosed with PC-1 when she was 3 months old. and callosities. The outcome was similar to the She presented thickened and discolored nails, previous technique. Thus, BTX injections have led focal palmoplantar keratoderma with painful blis- to substantial improvement in the patient’s qual- ters and erosions on the soles exacerbated by ity of life, allowing her retinoid treatment to be plantar hyperhidrosis (FIG. 1A), follicular hyper- withdrawn and enabling her to plan a pregnancy. keratosis on the extensor surfaces of the extrem- The treatment has shown no reduction in efficacy ities, oral and laryngeal leukokeratosis causing during 5 years of follow-up. vocal hoarseness. She had no family history of PC and was otherwise healthy. Foot blisters and cal- Case 2 losities were so painful that she was unable to The second patient, a 27-year-old man, presented walk and required the use of a wheelchair. At the focal plantar hyperkeratosis with associated hyper- age of 29, blood testing revealed a KRT6A mis- hidrosis, painful blisters on the soles and on the sense mutation (Dundee, Scotland), making the back of the feet (FIG. 2A), thickened nails and oral diagnosis of PC type 1 definitive. She had previ- leukokeratosis. The appearance of lesions began ously been treated with acitretin 25 mg/day, during infancy, and at the age of 23, a blood test which was later tapered to 10 mg/day, and kera- revealed a KRT6A missense mutation (Dundee, tolytic agents for more than 10 years with only Scotland). Lesions in palms were less severe. He partial effectiveness. In 2010, following informed had a son with the condition, although no formal consent, she was offered toxin plantar injections. genetic testing had been conducted on him. Kera- After paring down the callosities by curettage, we tolytic agents and oral retinoid treatment had pre- administered intradermal injections of 100 U of viously been applied with poor clinical response. botulinum toxin type A (BTX-A) (BotoxVR , Allergan, Five months ago we started treatment with BTX-A Inc, Irvine, CA) in each sole and 50 U in each at a dose of 150 U for each foot, concentrating the

2 Efficacy of botulinum toxin in pachyonychia congenita type 1

FIG. 2. Case 2 (A) Painful blisters and erosions on the back of the feet. (B) Improvement after BTX injections.

injections in the affected areas. A remarkable clini- PC showed a remarkable reduction in plantar cal improvement and reduction in plantar pain blistering and pain after BTX injections (9–11). was seen at Day 7 postinjection (FIG. 2B), with a BTX inhibits eccrine sweat glands by blocking sustained response thus far. He had previously the acetylcholine pathway, thus reducing hyper- been treated over an 18-month period at another hidrosis, maceration, and blistering of the fragile hospital with no preinjection curettage of foot epidermis. But even patients with no overt lesions and a total dose of 100 U, resulting in only hyperhidrosis or blisters reported pain relief. moderate improvement. This may be explained by a more direct effect of BTX on nociceptive C-fibers inhibiting neuro- peptide release from the sensory nerve axons. Discussion Also, BTX can decrease the release of pain medi- ators including substance P, calcium generelated Painful plantar hyperkeratosis is a very common peptide and glutamate, and also inhibit the vaso- feature in PC (95% of cases) and the most dis- dilation induced by the silent nociceptors that abling one, reducing quality of life due to plantar initiate neurogenic inflammation. These mecha- pain (1). Currently there is no specific and effec- nisms may explain the effect of BTX on tender- tive therapy for patients with PC, although certain ness in both keratinopathies. Activation of clinical trials have shown promising results. Tar- TRVP3 on keratinocytes leads to the release of geted therapeutic strategies including small inter- algogenic and pruritogenic substances inducing fering RNA, topical or systemic rapamycin or cutaneous pain. However, no reports have been simvastatin have recently been developed, but found to date regarding the effect of BTX-A in they cannot be offered on a regular basis (3–6). these new receptors (12–15). Many alternative treatments have previously been The technique applied was similar to treat- applied in this condition with variable and partial ment for axillary and palmar hyperhidrosis, but clinical response. These include: topical emol- the same outcome is achieved when concen- lients, keratolytic agents, mechanical removal of trating the entire dose directly below blisters excessive hyperkeratotic skin, avoidance of physi- and callosities. Mechanical removal of callosities cal activity, analgesia, and so forth (7). Among the by curettage prior to intradermal injections of systemic agents for treatment of PC, oral retinoid BTX-A may play a role in pain reduction due to therapy represents the treatment of choice (8). reduced friction. The optimal anaesthesia Epidermolysis bullosa simplex and PC are method may vary amongst different patients, congenital keratinopathies in which painful foot but superficial sedation with spontaneous venti- blisters and callosities severely reduce patient lation was applied in our cases. The dose of quality of life. Since 2006, a handful of studies BTX-A each session ranged from 75 to 150 U have begun to report the efficacy of BTX-A in per foot. Table 1 shows improvement scores treating both conditions. and our results in comparison with those of In 2010, a retrospective evaluation of 14 previous studies. The mean effect lasted 6 patients with epidermolysis bullosa simplex and months in both patients, no side effects were

3 Gonzalez-Ramos et al.

Table 1. Global outcome and details of therapy comparing our results with those of previous studies

Swartling and Vahlquist Swartling et al. Our cases (n 5 2) 2006 (9) (n 5 3) 2010 (11) (n 5 8) Type of BTX A (Botox) A (Dysport) 7 A (Dysport), 3 B (Neurobloc)* Dose (U) of 50 U (each palm) 125–300 U BTX-A: 200–350 BTX (range) 75–150 U (each foot) (each foot) (each foot) BTX-B: 2500 Improve in 1,5 u 2 callosities (average)† Improve in 3 u 1,6 blistering (average)† Improve in 2 u 2,1 pain (average)† DLQI score 23 (before)/3 (after) u u before/after treatment (average)‡ Effect duration 5–6 months (5, 5) 6 weeks– 2–12 months (3) (range and average) 6 months (3, 5) Number of treatments 1–9 (5) u 1–19 (8) (range and average) Follow-up (months) Case 1: 60; case 2: 5 24 60 Type of anaesthesia Superficial sedation IVRA§ 3 General, 5 IVRA Side effects None None 3/8 mild¶

*Due to treatment resistance, three patients were later switched to BTX B (Neurobloc). †0, no effect; 1, a little better; 2, much better; 3, very much better. ‡DLQI (Dermatology Life Quality Index): 0–1: no effect at all on patient’s life; 2–5: small effect on patient’s life; 6–10: moderate effect on patients life; 11–20: very large effect on patients life; 21–30 extremely large effect on patient’s life. §IVRA: intravenous regional anaesthesia. ¶Dysphagia in one case, and dry mouth, accommodation problems and urinary stress incontinence in two cases. n: number of patients. u: unknown. reported and effectiveness was sustained over Conflict of interests time. None of the authors have any conflict of interest to declare. Conclusion

There is, thus far no curative treatment for PC, and agents previously used to treat this condi- References tion have led to variable clinical outcomes. Until targeted therapeutic treatment can be offered on 1. Eliason MJ, Leachman SA, Feng BJ, Schwartz ME, Hansen CD. A review of the clinical phenotype of 254 a routine clinical basis, our findings support the patients with genetically confirmed pachyonychia con- usefulness of BTX-A plantar injections in provid- genita. J Am Acad Dermatol 2012: 67 (4): 680–686. ing sustained symptomatic relief to PC patients 2. Schnider P, Moraru E, Kittler H, et al. Treatment of focal and reducing painful foot blistering and callos- hyperhidrosis with botulinum toxin type A. long-term fol- ities with good tolerance and lasting effects. As low-up in 61 patients. Br J Dermatol 2001: 145: 289–293. 3. Chamcheu JC, Wood GS, Siddiqui IA, et al. Progress such, BTX-A treatment may offer a marked towards genetic and pharmacological therapies for kera- improvement in the quality of life for patients tin genodermatoses: current perspective and future affected by this rare and incapacitating disease. promise. Exp Dermatol 2012: 21 (7): 481–489 4. Hickerson RP, Flores MA, Leake D, et al. Use of self- delivery siRNAs to inhibit gene expression in an organo- Acknowledgements typic pachyonychia congenita model. J Invest Dermatol 2011: 131 (5): 1037–1044 The authors thank Savana Shakir and John Turcany 5. Hickerson RP, Leake D, Pho LN, Leachman SA, Kaspar RL. Rapamycin selectively inhibits expression of an inducible for their participation. keratin (K6a) in human keratinocytes and improves

4 Efficacy of botulinum toxin in pachyonychia congenita type 1

symptoms in pachyonychia congenita patients. J Dermatol 11. Swartling C, Karlqvist M, Hymnelius K, Weis J, Vahlquist Sci 2009: 56 (2): 82–88. A. Botulinum toxin in the treatment of sweat-worsened 6. Zhao Y, Gartner U, Smith FJ, McLean WH. Statins down- foot problems in patients with epidermolysis bullosa regulate K6a promoter activity: a possible therapeutic simplex and pachyonychia congenital. Br J Dermatol avenue for pachyonychia congenita. J Invest Dermatol 2010: 163 (5): 1072–1076 2011: 131 (5): 1045–1052 12. Bentsianov B, Zalvan C, Blitzer A. Noncosmetic uses of 7. Goldberg I, Fruchter D, Meilick A, Schwartz ME, Sprecher botulinum toxin. Clin Dermatol 2004: 22 (1): 82–88 E. Best treatment practices for pachyonychia congenita. 13. Kramer€ HH, Angerer C, Erbguth F, Schmelz M, Birklein F. J Eur Acad Dermatol Venereol 2014: 28 (3): 279–285. Botulinum Toxin A reduces neurogenic flare but has 8. Gruber R, Edlinger M, Kaspar RL, et al. An appraisal of almost no effect on pain and hyperalgesia in human skin. oral retinoids in the treatment of pachyonychia congen- J Neurol 2003: 250 (2): 188–193. ita. J Am Acad Dermatol 2012: 66 (6): e193–e199 14. Gazerani P, Pedersen NS, Staahl C, Drewes AM, Arendt- 9. Swartling C, Vahlquist A. Treatment of pachyonychia Nielsen L. Subcutaneous botulinum toxin type A reduces congenital with plantar injections of botulinum toxin. Br capsaicin-induced trigeminal pain and vasomotor reac- J Dermatol 2006: 154: 763–765 tions in human skin. Pain 2009: 141 (1–2): 60–69. 10. Abitbol RJ, Zhou LH. Treatment of epidermolysis bullosa 15. Nilius B, Bıro T, Owsianik G. TRVP3: time to decipher a simplex, Weber-Cockayne type, with botulinum toxin poorly understood family member! J Physiol 2014: 592 type A. Arch Dermatol 2009: 145 (1): 13–15 (Pt 2): 295–304

ORIGINAL ARTICLE

An appraisal of oral retinoids in the treatment of pachyonychia congenita

Robert Gruber, MD,a Michael Edlinger, MSc,b Roger L. Kaspar, PhD,c C. David Hansen, MD,d Sancy Leachman, MD, PhD,d Leonard M. Milstone, MD,e FrancesJ.D.Smith,PhD,f Alexis Sidoroff, MD,a PeterO.Fritsch,MD,a and Matthias Schmuth, MDa Innsbruck, Austria; Santa Cruz, California; Salt Lake City, Utah; New Haven, Connecticut; and Dundee, United Kingdom

Background: Pachyonychia congenita (PC), a rare autosomal-dominant keratin disorder caused by mutations in keratin genes KRT6A/B, KRT16,orKRT17, is characterized by painful plantar keratoderma and hypertrophic nail dystrophy. Available studies assessing oral retinoid treatment for PC are limited to a few case reports.

Objective: We sought to assess overall effectiveness, adverse effects, and patient perspective in patients with PC receiving oral retinoids.

Methods: In a questionnaire-based retrospective cross-sectional survey of 30 patient with PC assessing oral retinoids (10-50 mg/d for 1-240 months), we determined the clinical score, satisfaction score, visual analog pain scale, and adverse effects.

Results: In 50% of patients there was thinning of hyperkeratoses (average improvement 1.6 on a scale from e3to13) (95% conﬁdence interval 1.2-1.9, P \ .001). In all, 14% observed amelioration of their pachyonychia; 79% did not experience any nail change. The self-reported overall satisfaction score with oral retinoid treatment was 2 or greater in 50% of the patients (mean 4.5 on a scale of 1-10). Although 33% reported decreased and 27% increased plantar pain with treatment, 40% did not notice any pain change. All patients experienced adverse effects, and 83% reported to have discontinued medication. Risk/benefit analysis favored lower retinoid doses ( # 25 mg/d) over a longer time period ([5 months), compared with higher doses ([25 mg/d) for a shorter time ( # 5 months).

Limitations: The retrospective, cross-sectional study design is prone to a recall bias.

Conclusion: Oral retinoids are effective in some patients with PC. However, many patients discontinued medication because adverse effects outweighed the beneﬁts. Careful dose titration is warranted in patients informed about potential adverse effects. ( J Am Acad Dermatol 10.1016/j.jaad.2011.02.003.)

Key words: keratins; keratoderma; oral retinoids; pachyonychia congenita.

achyonychia congenita (PC) (Online extremely painful diffuse or focal symmetrical hy- Mendelian Inheritance in Man [OMIM] perkeratosis of palms and soles sometimes associated P #167200 for PC-1, #167210 for PC-2) is a rare with erosions, follicular keratosis on the extensor autosomal-dominant ectodermal dysplasia that is surfaces of the extremities, oral leukokeratosis, and characterized by hypertrophic nail dystrophy, other ectodermal defects.1-3 At the molecular level,

From the Department of Dermatologya and Department of Accepted for publication February 1, 2011. Medical Statistics, Informatics and Health Economics,b Inns- Reprint requests: Matthias Schmuth, MD, Department of bruck Medical University; TransDerm Inc, Santa Cruzc; Depart- Dermatology, Innsbruck Medical University, Anichstr. 35, 6020 ment of Dermatology, University of Utahd; Department of Innsbruck, Austria. E-mail: [email protected]. Dermatology, Yale University School of Medicine, New Havene; Published online May 21, 2011. and College of Life Science, University of Dundee.f 0190-9622/$36.00 Supported in part by the International Pachyonychia Congenita Ó 2011 by the American Academy of Dermatology, Inc. Research RegistryePachyonychia Congenita Project. doi:10.1016/j.jaad.2011.02.003 Conflicts of interest: None declared.

e1 e2 Gruber et al JAM ACAD DERMATOL

PC is caused by dominant-negative mutations in included in this questionnaire-based retrospec- keratin genes KRT6A, KRT6B, KRT16, and tive cross-sectional study. The study was con- KRT17.4-6 As these keratins are expressed in differ- ducted in accordance with the principles of the entiated epithelial structures such as the nail bed, Declaration of Helsinki and written informed palmoplantar epidermis, and the oral mucosa, these consent was obtained from all 30 patients before are the affected tissues in PC-1.2,5 The hitherto com- enrollment. mon division of PC into PC-1 and PC-2 subtypes according to the clinical pre- Questionnaire sentation is being increas- Questionnaire-based pa- CAPSULE SUMMARY ingly replaced with tient scoring was used to genotype-inclusive nomen- evaluate clinical score, satis- d Available studies assessing oral retinoid clature (eg, PC-K6a, PC-K6b, faction score, visual analog treatment for pachyonychia congenita PC-K16, and PC-K17).7 pain scale, and adverse ef- (PC), a rare keratin disorder characterized Treatment of PC is notori- fects of treatment with oral by painful plantar keratoderma and nail ously difﬁcult. Because PC is retinoids by self-assessment. dystrophy, are limited to a few case rare (;1:500,000-1:1000,000), Patients were either inter- reports. available studies assessing viewed in person or via tele- therapeutic regimen are lim- d Our study shows that although phone. The questionnaire ited to a few case reports and treatment of PC with oral retinoids is items are summarized in case series. Basic measures effective in some individuals with PC, Table II. include topical emollients, increased pain is a common adverse keratolytic agents, mechanical effect. Statistics removal of excessive hyper- d The study presents the benefit/risk ratio All data were analyzed keratotic skin, and avoidance of oral retinoids in PC and provides new with software (SPSS, Version of physical activity. Among insight in favorable dosing regimens. 17.0 for Windows, SPSS Inc, the systemic agents for treat- Chicago, IL). Statistical differ- ment of PC, the best results ences between groups were have been reported with oral administration of vitamin determined by using the Mann-Whitney U test with 8-11 A derivatives, ie, retinoids. However, the evidence signiﬁcance conferred when P less than .05. To for their effectiveness is based on anecdotal reports assess predictors of the effectiveness of drug treat- and no systematic retrospective or prospective studies ment we estimated odds ratios and 95% confidence are available. For severe inherited disorders of corni- intervals (CIs) with logistic regression modeling; this fication such as ichthyoses and psoriasis, oral retinoid analysis was restricted to the retinoids acitretin and 12,13 therapy represents the treatment of choice. In PC, isotretinoin given the small number of cases on other this therapeutic approach is particularly attractive, retinoids. because in screening assays, retinoids have been noted to suppress mutant keratin expression (W.H. Irwin RESULTS McLean, DSc, FRSE, oral communication, May 2010). Natural course of plantar hyperkeratoses in PC We here present a questionnaire-based retrospective As PC is a dynamic disease, 25 of the 30 patients cross-sectional survey of 30 patients assessing effec- (83%) (Table I) who received oral retinoids reported tiveness, adverse effects, and overall patient satisfac- spontaneous changes in plantar hyperkeratoses tion of oral retinoid therapy for PC. Our goal was to while not using any medication. On a scale of e3 establish the benefit/risk ratio, identify favorable dos- (much worse) to 13 (much better) plantar thickening regimens, and determine if a future prospective ing was reported to range between e1.8 and 0.9, trial for oral retinoid treatment of PC is justified. with an average change of e0.45, ie, the majority of patients reported worsening of the disease while not METHODS taking any drugs. These results indicate that without Patients medication plantar hyperkeratoses in PC varies over All individuals presenting with PC with known time with little spontaneous improvement. mutations in KRT6a, KRT16,orKRT17 who were enrolled in the International Pachyonychia Treatment effectiveness Congenita Research Registry between 2004 and Fifteen patients (50%) reported decreased plantar 2010 and previously treated with oral retinoids hyperkeratoses, ie, thinning of calluses, when taking (acitretin, etretinate, isotretinoin, or vitamin A) at medication (Fig 1). On a scale from e3 (much worse) different doses and durations (Table I)were to 13 (complete improvement) the average JAM ACAD DERMATOL Gruber et al e3

Table I. Patient demographics and retinoid therapy

No. Gene Mutation Sex Retinoid Dose, mg/d Duration, mo 1 K6a D432_E470dup F Isotretinoin NA 1 2 K16 N125D M Isotretinoin 25 24 3 K6a N172del M Etretinate 50 6 4 K16 L132P F Acitretin NA 6 5 K6a N171K F Isotretinoin NA 3 6 K17 N92S F Isotretinoin 50 3 7 K16 L132P F Etretinate 12,5 240 8 K16 L132P F Acitretin 50 12 9 K6a L468P F Acitretin 10 24 10 K6a L469R F Isotretinoin NA 6 11 K16 L132P M Acitretin NA 12 12 K6a N171Y M Acitretin 25 60 13 K16 S130del M Isotretinoin 40 1 14 K17 N92S F Isotretinoin 40 6 15 K16 L132P F Isotretinoin 40 6 16 K6a N172del F Acitretin 25 4 17 K6a E461K M Isotretinoin NA 3 18 K16 K15X F Isotretinoin NA 3 19 K17 M88T F Isotretinoin 40 9 20 K17 M88T M Acitretin 25 24 21 K17 N92S M Isotretinoin NA 3 22 K16 L124H F Acitretin 30 3 23 K16 N125S M Acitretin 25 3 24 K6a L468P F Isotretinoin 40 6 25 K17 N92S F Isotretinoin NA 2 26 K16 L132P M Acitretin 25 5 27 K16 N125S M Acitretin 20 4 28 K6a N172del M Etretinate 10 2 29 K6a N172del M Acitretin 25 12 30 K16 L132P F Ro-A-Vit 50 1

F, Female; M, male; NA, not applicable. improvement was 1.6 (95% CI 1.2-1.9, P \ .001). dreadful pain. Decreased plantar pain during ther- Although 4 of the 28 patients (14%) for whom data on apy was reported in 10 patients (33%) whereas nail thickening were available reported amelioration 8 patients (27%) experienced increased pain and of their pachyonychia (thinning of nails, lighter the remaining 12 patients (40%) did not report any color) with an average improvement of 1.5 on a change in their pain. scale from e3to13, the majority of the individuals, In the majority of individuals with decreased pain, ie, 22 (79%), did not experience any change in nail the improvement occurred within the ﬁrst 3 to 4 involvement and only two patients (7%) reported weeks of taking oral retinoids. In the 10 patients with worsening of their pachyonychia. The self-reported decreased plantar pain, the degree of pain amelio- overall satisfaction score with oral retinoid treatment ration ranged from 1 to 7 on a scale of 1 (minimal) to was greater than or equal to 2 in 15 patients (50%), 10 (most), with a mean change of 3.4 (95% CI 1.8-5.1, with a mean of 4.5 on a scale of 1 (lowest) to 10 P = .001). In the 8 patients who experienced in- (highest). Notably, only 7 patients (23%) recommend creased pain, the degree of pain worsening ranged the use of oral retinoids to others and even fewer, 5 from 3 to 10, with a mean change of 7.1 (95% CI patients (17%), are still using the medication. 5.4-8.8, P \.001).

Effects on pain Retinoid dosing Before treatment with oral retinoids, the overall Among the participants of this study taking oral pain when walking was quantiﬁed by a visual analog retinoids the dose ranged from 10 to 50 mg/d pain scale (0 meaning no pain and 10 the worst pain (Table I). Forty percent of patients who were treated ever experienced) and reported as 6, which signiﬁed with doses of more than 25 mg/d (higher doses), e4 Gruber et al JAM ACAD DERMATOL

Table II. Questionnaire

Natural course of PC without treatment d Did your PC condition change at times when not using any drug treatment? d If yes, indicate range of change on scale of e3 (much worse) to 13 (much better), with 0 being your normal condition Effectiveness/satisfaction d Was drug effective? d What was your overall satisfaction on scale of 1-10 (1 = lowest, 10 = highest)? d Are you still using drug? d Would you recommend drug to others? Clinical treatment response d While taking drug did your PC calluses/nails change? d If yes, did calluses/nails get better or worse? d If calluses/nails got better, what was improvement? d What was clinical improvement/deterioration in your PC calluses/nails on scale of e3 (much worse) to 13 (complete improvement)? Pain d On scale of 0 (no pain) to 10 (worst pain you have ever experienced), rate your overall pain before taking drug d While on drug, did you notice change in pain? d If yes, did change cause more/less pain? d If yes, indicate when change occurred d If yes, mark change in your pain using scale of 1 (minimal) to 10 (most) Dose and duration d Which oral retinoid did you use (trade name)? d If you remember, what dose or strength of medication did you use? d How long did you take this medication? d Was dose of your retinoid medication changed during treatment? d If yes, to higher/lower dose? d If yes, why was dose changed? d Did change give more improvement? d Did change create more adverse effects? Adverse effects d Did you experience any adverse effects? d Did you experience any dry eyes/dry lips/dry skin/peeling of skin/hair loss/headaches/bone or joint pain/other adverse effects while taking drug? d Did your doctor find any liver problems caused by use of retinoid treatment? d Did you think improvement you experienced while taking drug was sufficient that you wanted to continue in spite of any adverse effects? d Based on your experience, would you be interested in using oral retinoids again?

PC, Pachyonychia congenita. reported overall effectiveness and a mean overall Treatment duration satisfaction score of 1.6 on a scale from 1 (lowest) to The duration of treatment with oral retinoids 10 (highest). In comparison, 73% of patients receiv- ranged from 1 to 240 months, with 50% of the ing oral retinoid doses of less than or equal to patients receiving therapy for longer than 5 months 25 mg/d (lower doses) reported overall effectiveness (longer duration) and 50% less than or equal to 5 (P = .14) and a mean overall satisfaction score of months (shorter duration) (Table I). Of patients who 4.2 on a scale from 1 to 10 (P = .02). These results were treated for a longer duration, 67% reported suggest that although lower doses are not signifi- overall effectiveness and a mean overall satisfaction cantly different in their effectiveness, the ratios score of 3.6 on a scale from 1 to 10 compared with among effectiveness, pain, and adverse effects 33% effectiveness (P = .07) and a score of 1.9 (P = .06) were more favorable with lower doses. Decreased in the patients receiving oral retinoids for a shorter plantar pain during therapy with oral retinoids was duration. An improvement of PC calluses was reported in 50% treated with higher doses and 67% of reported in 83% of patients treated longer than patients treated with lower doses (P = .53), ie, higher 5 months and 75% treated for 5 months or less (P = retinoid doses were not superior in reducing pain. .66); the overall change in plantar hyperkeratosis on JAM ACAD DERMATOL Gruber et al e5

Fig 1. A, Thick plantar hyperkeratosis in patient with pachyonychia congenita before treatment with oral retinoids. B, Thinning of calluses while on therapy with acitretin. However, medication had to be discontinued because of adverse effects, ie, peeling of skin and increased pain and vulnerability of feet when walking. a scale from e3 (much worse) to 13 (complete Adverse effects improvement) was 0.7 for the longer duration group All study patients experienced adverse effects. versus 0.3 for the shorter duration group (P = .38). With the exception of one man all patients reported A decrease in plantar pain was reported in 67% of dry lips, 15 (50%) dry eyes, 27 (90%) dry skin, 11 patients treated for a longer period and 33% treated (37%) peeling of the skin, 9 (30%) hair loss, 6 (20%) for a shorter duration, but this again was not statis- headaches, 5 (17%) bone or joint pain, 4 (13%) sun tically different (P = .19). Interestingly, the percent- sensitivity, and one patient depression, fatigue, and age of patients with increased plantar pain was lower developing of bone spurs, respectively. In one pain the group with longer treatment duration (33% tient treatment with oral retinoids was stopped vs 67%). because of liver enzyme abnormalities. Considering retinoid doses, ie, more than 25 mg/d versus less than or equal to 25 mg/d, the Retinoid classes prevalence of adverse effects such as dry lips (90% vs Only patients who had received acitretin (N = 12) 100%, P = .29), dry eyes (20% vs 64%, P = .05), dry or isotretinoin (N = 14) were further compared skin (90% vs 91%, P = .94), skin peeling (50% vs 36%, because of the small number of patients treated P = .54), hair loss (20% vs 55%, P = .11), headaches with etretinate or vitamin A (Table I). The overall (20% vs 18%, P = .92), and bone/joint pain (20% vs effectiveness was 58% for acitretin and 36% for 36%, P = .42) was comparable. There was no isotretinoin (P = .26), the overall satisfaction score difference in adverse effects relative to treatment was 3.5 for acitretin versus 2.1 for isotretinoin (P = duration and there was also no significant difference .14). Ninety percent of patients in the acitretin in adverse effects when comparing acitretin with group compared with 57% in the isotretinoin group isotretinoin. reported thinning of calluses (P = .13) (Fig 1); the Nine patients (30%) decided to continue treat- overall change on a scale from e3 (much worse) to ment with oral retinoids despite adverse effects 13 (complete improvement) was 0.9 versus e0.1 because they perceived PC improvement as sufﬁ- (P = .05). In all, 63% versus 50% experienced cient. Based on their experience, 14 patients (47%) decreased plantar pain while on acitretin and would be interested in using oral retinoids again, isotretinoin, respectively (P = .65). Although not primarily because of the overall effectiveness they significantly different, these data indicate that aci- experienced. Sixteen patients (53%) would not use tretin may have a slight edge over isotretinoin in oral retinoids again. However, the question used for treating PC. this assessment (Table II) did not address in detail e6 Gruber et al JAM ACAD DERMATOL

why patients stated that they would or would not improvement of hereditary epidermolytic palmo- want to use oral retinoids again. plantar keratoderma in 4 patients treated with an oral aromatic retinoid for 5 months, resulting in Predictors of effectiveness normal-appearing skin. However, therapy had to be To identify additional patient subsets that might be discontinued as the vulnerability and sensitivity more likely to benefit from oral retinoid treatment, restricted normal function of hands and feet. In the we used logistic regression modeling. The calculated current study roughly only one third of patients odds ratios were 0.13 for female versus male (95% CI experienced improvement of pain although 50% of 0.02-0.89), 1.6 for acitretin versus isotretinoin (95% CI the patients reported improved plantar hyperkera- 0.3-9.6), and 0.7 for age, per 10 years (95% CI 0.4- tosis, ie, in some cases even though calluses thinned, 1.5). These results indicate that neither patient age there was increased pain. nor retinoid type were predictors of effectiveness. In Recently, mutations in KRT6, KRT16, and KRT17 contrast, a benefit from oral retinoid treatment was were correlated with characteristic clinical findings less likely in female than in male patients. in patients with PC; ie, KRT6B was associated with increased pain intensity.18 When stratifying our pa- DISCUSSION tients by genotypes, because of the small subgroups, The few available case reports and case series no further analysis regarding treatment effectiveness assessing therapy of PC with oral retinoids yielded was possible. contradictory results with a tendency toward more This study has several limitations including its favorable outcomes with treatment. Dupreetal 8 retrospective, cross-sectional study design, which is asserted improvement of PC calluses and decreased prone to a recall bias. Although the measurements pain in 3 patients treated with an aromatic retinoid, were patient-based and subjective, ie, not assessed Hoting and Wassilew9 reported a remission of palmo- by a physician, the study end points are patient- plantar hyperkeratosis but no changes in pachyo- centered in a positive sense, in that they should be nychia in two patients treated with etretinate 75 mg/d highly relevant for reflecting patient perception of and a relapse when reducing the drug to 30 mg/d after treatment. The lack of laboratory monitoring for several months, Carabott et al10 described a patient potential adverse effects such as liver function testing who experienced reduced plantar hyperkeratosis may result in an overestimation of the benefit/risk after 3 months of etretinate therapy at a dose of ratio of the treatment modality, ie, our study design is 50 mg/d, and Lim et al11 reported amelioration of biased to exclude patients with severe adverse calluses but not of nail changes in a patient treated effects. In the current study, discontinuation of oral with 30 mg acitretin daily. In contrast, two additional retinoid therapy was only necessary in one patient case reports did not show any clinical benefit for oral because of elevated liver transaminases. This is in retinoids in the treatment of PC. Thomas et al14 accordance with two previous studies, in which reported a father and his son who did not show similar retinoid doses have been used for the treat- improvement of plantar keratoderma despite therapy ment of various forms of ichthyoses and psoriasis, with high doses of isotretinoin. Similarly, Soyuer and and no severe adverse effects were reported.12,13 Candan15 described failure of etretinate in a child This study demonstrates a potential advantage of with PC, but the treatment was only administered for 5 treatment with lower doses of acitretin for a longer weeks, and the dose had to be progressively lowered duration compared with therapy with higher doses, because of hypertriglyceridemia. Consistent with the shorter duration, and isotretinoin. Recently it was contradictory findings in these published reports, shown that in the treatment of patients with psoriasis in the current study involving 30 patients from the low-dose acitretin (25 mg/d) was associated with PC registry, oral retinoids resulted in thinning of fewer common adverse effects than high-dose aci- calluses (decrease in hyperkeratosis) in only a subset tretin (50 mg/d).13 Because lower doses may have of study patients. a better risk/benefit ratio, it might be beneficial A very important aspect of PC treatment is that to begin treatment at a lower dose (eg, acitretin thinning of calluses does not necessarily imply de- 10-25 mg/d) with further dose adjustments based on creased plantar pain when walking, as reported for patient’s response. Alternatively, treatment may be other types of palmoplantar keratoderma. In a family initiated at a higher dose and subsequently adjusted with keratoderma of the soles associated with blis- depending on pain and adverse effects. It is impor- tering but lack of pachyonychia, treatment with tant that patients are fully informed about potential isotretinoin resulted in callus reduction, but blister- adverse effects before initiation of therapy including ing worsened and pain increased.16 This was also the possibility of increased pain when on oral observed by Fritsch et al,17 who described a dramatic retinoids. JAM ACAD DERMATOL Gruber et al e7

In conclusion, the results of our study confirm that 7. Eliason MJ. A review of the clinical phenotype in patients with pachyonychia congenita. Submitted for publication. treatment of PC with oral retinoids is effective in some individuals with PC. Randomized, controlled, 8. Dupre A, Christol B, Bonafe JL, Touron P. Pachyonychia congenita: three familial cases; effects of the treatment by aromatic prospective clinical trials with both objective and retinoid (RO 10.9359). Ann Dermatol Venereol 1981;108:145-9. patient-centered subjective end points are warranted 9. Hoting E, Wassilew SW. Systemic retinoid therapy with etret- to further define the patient subsets that most benefit inate in pachyonychia congenita [in German]. Hautarzt 1985; from this treatment option. 36:526-8. We are indebted to the participating patients, and 10. Carabott F, Archer CB, Griffiths WA. Etretinate-responsive pachyonychia congenita. Br J Dermatol 1988;119:551-3. to Mary Schwartz and the members of the PC Project 11. Lim TW, Paik JH, Kim NI. A case of pachyonychia congenita Medical Scientific Advisory Board for their valuable with oral leukoplakia and steatocystoma multiplex. J Dermatol advice. 1999;26:677-81. 12. Verfaille CJ, Vanhoutte FP, Blanchet-Bardon C, van Steensel MA, Steijlen PM. Oral liarozole vs acitretin in the treatment of REFERENCES ichthyosis: a phase II/III multicenter, double-blind, random- 1. Feinstein A, Friedman J, Schewach-Millet M. Pachyonychia ized, active-controlled study. Br J Dermatol 2007;156:965-73. congenita. J Am Acad Dermatol 1988;19:705-11. 13. Pearce DJ, Klinger S, Ziel KK, Murad EJ, Rowell R, Feldman SR. 2. Leachman SA, Kaspar RL, Fleckman P, Florell SR, Smith FJ, Low-dose acitretin is associated with fewer adverse events McLean WH, et al. Clinical and pathological features of than high-dose acitretin in the treatment of psoriasis. Arch pachyonychia congenita. J Investig Dermatol Symp Proc Dermatol 2006;142:1000-4. 2005;10:3-17. 14. Thomas DR, Jorizzo JL, Brysk MM, Tschen JA, Miller J, Tschen 3. Jackson AD, Lawler SD. Pachyonychia congenita; a report of EH. Pachyonychia congenita: electron microscopic and epi- six cases in one family, with a note on linkage data. Ann Eugen dermal glycoprotein assessment before and during isotreti- 1951;16:142-6. noin treatment. Arch Dermatol 1984;120:1475-9. 4. Smith FJ, Jonkman MF, van Goor H, Coleman CM, Covello SP, 15. Soyuer U, Candan MF. Failure of etretinate therapy in pach- Uitto J, et al. A mutation in human keratin K6b produces a yonychia congenita. Br J Dermatol 1987;117:264. phenocopy of the K17 disorder pachyonychia congenita type 16. Baden HP, Bronstein BR, Rand RE. Hereditary callosities with 2. Hum Mol Genet 1998;7:1143-8. blisters: report of a family and review. J Am Acad Dermatol 5. McLean WH, Rugg EL, Lunny DP, Morley SM, Lane EB, 1984;11:409-15. Swensson O, et al. Keratin 16 and keratin 17 mutations cause 17. Fritsch P, Honigsmann H, Jaschke E. Epidermolytic hereditary pachyonychia congenita. Nat Genet 1995;9:273-8. palmoplantar keratoderma: report of a family and treatment 6. Bowden PE, Haley JL, Kansky A, Rothnagel JA, Jones DO, with an oral aromatic retinoid. Br J Dermatol 1978;99:561-8. Turner RJ. Mutation of a type II keratin gene (K6a) in 18. Eliason MJ. IPCRR Data 2010. Paper presented at: International pachyonychia congenita. Nat Genet 1995;10:363-5. PC Consortium Symposium; May 2010, Atlanta, GA.

First-in-human Mutation-targeted siRNA Phase Ib Trial of an Inherited Skin Disorder

Sancy A Leachman1, Robyn P Hickerson2, Mary E Schwartz3, Emily E Bullough1, Stephen L Hutcherson4, Kenneth M Boucher1, C David Hansen1, Mark J Eliason1, G Susan Srivatsa5, Douglas J Kornbrust6, Frances JD Smith7, WH Irwin McLean7, Leonard M Milstone8 and Roger L Kaspar2

1Department of Dermatology Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah, USA; 2TransDerm Inc., Santa Cruz, California, USA; 3PC Project, Salt Lake City, Utah, USA; 4Visionary Therapeutics Corporation, Richmond, Virginia, USA; 5ElixinPharma, Encinitas, California, USA; 6Preclin- sight, Reno, Nevada, USA; 7Division of Molecular Medicine, Medical Sciences Institute, University of Dundee, Dundee, UK; 8Department of Dermatology, Yale University, New Haven, Connecticut, USA

The rare skin disorder pachyonychia congenita (PC) is virus.6 This phase I siRNA trial for respiratory syncytial virus an autosomal dominant syndrome that includes a dis- demonstrated few side effects and promising signs of efficacy. The abling plantar keratoderma for which no satisfactory siRNA trial reported herein is the first-in-man siRNA trial for a treatment is currently available. We have completed a skin disorder as well as the first to target a mutated gene causing phase Ib clinical trial for treatment of PC utilizing the first an inherited disease. short-interfering RNA (siRNA)-based therapeutic for skin. Pachyonychia congenita (PC) is an ultrarare (less than a few This siRNA, called TD101, specifically and potently tar- thousand cases worldwide7), highly disabling, autosomal domi- gets the keratin 6a (K6a) N171K mutant mRNA without nant inherited disorder that affects the nails, skin, oral mucosae, affecting wild-type K6a mRNA. The safety and efficacy of hair, and teeth.8,9 Manifestations are regionally variable but focally TD101 was tested in a single-patient 17-week, prospec- persistent and do not spontaneously resolve. The most disabling tive, double-blind, split-body, vehicle-controlled, dose- feature of PC is painful plantar calluses for which no satisfactory escalation trial. Randomly assigned solutions of TD101 treatment is currently available. Although PC symptoms and pain or vehicle control were injected in symmetric plantar levels vary between patients, many are unable to walk without the calluses on opposite feet. No adverse events occurred aid of crutches at least intermittently or must use a wheelchair. PC during the trial or in the 3-month washout period. Sub- patients often walk on their knees while at home to avoid contact jective patient assessment and physician clinical efficacy with the plantar calluses. Although it is not possible to effectively or measures revealed regression of callus on the siRNA- safely remove PC calluses completely, patients routinely use pumice treated, but not on the vehicle-treated foot. This trial stones and razor blades to groom the calluses to help alleviate pain. represents the first time that siRNA has been used in a These calluses are so exquisitely sensitive that patients rarely allow clinical setting to target a mutant gene or a genetic dis- others, even family members, to touch their feet. PC patients could order, and the first use of siRNA in human skin. The cal- benefit enormously from an effective, locally administered therapy. lus regression seen on the patient’s siRNA-treated foot PC is caused by mutations in either keratin K6, K16, or K17 appears sufficiently promising to warrant additional that act by a dominant-negative mechanism to cause the disease 8,9 studies of siRNA in this and other dominant-negative symptoms. Selective depletion of the mutated keratin has the skin diseases. potential to directly target the molecular etiology of the disease, and there is compelling evidence from animal models, at least in Received 19 June 2009; accepted 19 October 2009; advance online the case of a similar keratin disorder, epidermolysis bullosa simplex publication 24 November 2009. doi:10.1038/mt.2009.273 (due to keratin 14), that even partial reduction of mutant keratin expression may have a beneficial clinical effect.10 Further, substan- Introduction tial redundancy among PC-related keratins, as demonstrated in The emergence of short-interfering RNA (siRNA) as a powerful knockout mice,11 reduces the likelihood of a severe adverse event tool to reduce target gene expression in vitro and in animal model if a single keratin is eliminated (e.g., if both wild-type and mutant systems has led to recent therapeutic trials of siRNAs for macular keratin 6a (K6a) were inhibited by the siRNA). Overall, the focal degeneration, diabetic macular edema, solid tumors, respiratory nature of PC, our understanding of the underlying molecular syncytial virus, hepatitis B, and human immunodeficiency viral defects of the disorder, the lack of effective PC therapies, and the infections.1–5 However, these trials have not yet been published. ability to visually observe changes during treatment make PC a To date, the only published clinical trial reporting the use of a particularly good human skin disease model for testing siRNA in siRNA is a safety and tolerability study for respiratory syncytial a proof-of-concept trial for genetic disorders.

Correspondence: Sancy A Leachman, Huntsman Cancer Institute, Department of Dermatology, 2000 Circle of Hope, Salt Lake City, Utah 84112-5550, USA. E-mail: [email protected]

Molecular Therapy 1 © The American Society of Gene & Cell Therapy siRNA Phase Ib Trial of an Inherited Skin Disorder

Preclinical studies have demonstrated that the TD101 siRNA schedule). Efficacy and safety measurements are detailed in the is safe, as well as highly potent and specific.12,13 This siRNA has Materials and Methods section, as well as the Supplementary been shown to specifically target the cytosine-to-adenine single Materials and Methods. nucleotide K6a mutation (resulting in the amino-acid change N171K) in patient-derived immortalized keratinocytes.13 It has Results also been shown to reverse the mutant phenotype of cells in a At the conclusion of the washout period, the blinding code was dominant-negative tissue culture model by restoring their abil- broken and revealed that the right foot had received TD101 ity to form a structurally intact keratin intermediate filament siRNA, whereas the left foot had received the vehicle-control solu- network.12,13 Furthermore, the TD101 siRNA has been tested in tion. Both subjective patient data and physician-derived clinical a mouse model using bicistronic reporter constructs consisting data suggest that in similar symmetric calluses, there was a posi- of firefly luciferase linked to either wild-type or N171K mutant tive effect of the injection of TD101 in the right foot, but not in the K6a. Co-delivery of these constructs and the mutation-specific vehicle control–injected left foot. siRNA resulted in potent inhibition of the mutant (but not the During the first 2 months of the trial, no dramatic differences wild-type) version of the gene, as assayed by in vivo biolumi- (subjective or objective) between feet were noted by either the nescence imaging12 (and unpublished results). In a compari- patient or physician. At this point in the trial, there were no visible son of unmodified and modified siRNA, we found that some responses in the calluses of either foot that would indicate either modifications eliminated the single nucleotide specificity, significant injury or efficacy from the drug. At approximately day and no increase in efficacy was observed using modified -ver 70 of the trial (dose = 2 ml; 3 mg/ml), the patient’s subjective eval- sions (data not shown). Furthermore, we reasoned that if any uation of the injected callus (“If you are receiving a study medica- unmodified siRNA were to enter the bloodstream, it would be tion, evaluate if it is working and improving your PC symptoms.” quickly degraded, increasing the safety profile. For these rea- 0 = definitely working; 10 = definitely not working) began to indi- sons, unmodified TD101 was used in this trial. A mouse toxicity cate a marked difference in the right foot, but no change in the left study demonstrated a lack of serious toxicity when the TD101 foot (Figure 1). Measurements of the injected calluses also began siRNA was delivered at high-dose levels by intradermal injec- to show a statistically significant decrease in length of the callus on tion.13 Thus, preclinical studies in in vitro and in vivo model the right foot only (Figure 2). On day 98 of the trial, after dose 28 systems demonstrated both safety and effective inhibition of (dose = 2 ml; 5.0 mg/ml), the callus at the site of injection on the N171K K6a by the TD101 siRNA. right foot began to fall away and revealed healthy, pink skin. The The clinical efficacy and safety of TD101, administered by underlying skin was remarkably nontender to palpation, whereas intralesional injection into a plantar callus, was evaluated in surrounding areas of callus retained sensitivity (Figure 3a). This a single patient using a prospective, double-blind, split-body, type of behavior had never been observed by the patient previously vehicle-controlled, dose-escalation study design. Treatment was (Figure 1 and patient personal communication). This reduction completed after 17 weeks of twice-weekly injections and was in tenderness is the most dramatic clinical observation in the trial. followed by a 3-month washout period (see Table 1 for dosing By day 115, not only was the reduced length of the callus obvious,

Table 1 dose-escalation schedule Concentration of Total dose Week Dose no. Days Volume (ml) TD101 (mg/ml) TD101 (mg) 1 1–2 1–7 0.1 1.0 0.10 2 3–4 8–14 0.25 1.0 0.25 3 5–6 15–21 0.50 1.0 0.50 4 7–8 22–28 1.0 1.0 1.0 5 9–10 29–35 1.5 1.0 1.5 6 11–12 36–42 2.0 1.0 2.0 7 13–14 43–49 2.0 1.5 3.0 8 15–16 50–56 2.0 2.0 4.0 9 17–18 57–63 2.0 2.5 5.0 10 19–20 64–70 2.0 3.0 6.0 11 21–22 71–77 2.0 3.5 7.0 12 23–24 78–84 2.0 4.0 8.0 13 25–26 85–91 2.0 4.5 9.0 14 27–28 92–98 2.0 5.0 10.0 15 29–30 99–105 2.0 6.0 12.0 16 31–32 106–112 2.0 7.0 14.0 17 33 113–119 2.0 8.5 17.0

2 www.moleculartherapy.org © The American Society of Gene & Cell Therapy siRNA Phase Ib Trial of an Inherited Skin Disorder

110 20 10 20

100 15 8 15

90 10 6 10 Dose (mg) Dose (mg) Callus length (nm) Improvement score 80 5 4 5

Vehicle Vehicle TD101 TD101 70 Dose 0 2 Dose 0 −50 050 100 150 200 −50 050 100 150 200 Study day Study day Figure 2 the length of the callus on the right foot (TD101, blue), Figure 1 Patient assessment demonstrates subjective improvement but not the left foot (vehicle, red), decreased significantly during in the right foot (TD101, blue) but not the left foot (vehicle, red) the dosing period (day 0 to day 118) (drug versus vehicle, P = 0.004). as determined by daily diary entries. Note that the scores for the left The measured callus length (dots and open circles) and a piecewise lin- foot have been slightly offset so that the left and right foot scores do ear fit through the callus lengths are both plotted (see Materials and not overlap at baseline. The improvement scores and a curve fit through Methods). The fit has a change in slope at the end of the dosing period. the scores are both plotted. The corresponding dose (gray) is indicated The corresponding dose is indicated by a separate graph. by a separate graph. An improvement score of 10 represents “definitely not working,” whereas a score of 0 represents “definitely working.” The improvement returns to baseline after treatment is completed. obtain biopsy tissue for measurement of allele-specific mRNA levels because the protocol required repeated local adminis- but there was also an area clinically free of hyperkeratosis in the tration of drug, and the safety evaluation arm of our protocol center of the injection site (Figure 3a). Subjective and objective might have been compromised by repeated biopsies of the foot changes in the right foot began returning toward baseline after the in an individual prone to blisters and infection. Recent data drug was discontinued and reached baseline ~30–50 days after the from animal studies of a siRNA for macular degeneration sug- last dose. Figure 2 shows that the trend to shortening of the cal- gest that some of the clinical response may be a nonspecific lus length occurred prior to day 98 when the callus began peeling reaction to the siRNA.14 Once an animal model of PC is avail- away from the injection site. able, scrambled siRNAs and tissue biopsies will be important Because of the dynamic nature of callus development and res- mechanistic controls for future studies. However, there is no olution, it is unclear at what dose the callus first began to regress. evidence from our preclinical studies, where nonspecific siRNA The cellular turnover and retention in PC calluses have never been controls were used at every stage, that TD101 is acting in a non- measured, so it is difficult to surmise whether the lower doses of specific manner unrelated to selective degradation of mutant TD101 began to have a cumulative effect that was not observed K6a mRNA.12,13,15 until later or if the response at later time-points was due to an The degree of pain experienced by the patient at the time of increased dose at that time. Future investigation with a single- injection is a significant concern. Although pain related to the dose level over a prolonged period of time may help to address injection did not persist longer than a few hours after injection, these questions. the intense pain experienced at the time of injection will limit There were no clinical signs of a systemic response to the the utility of the drug by this delivery method. At the incep- TD101 injections (i.e., outside the injection site) in the skin, tion of this trial, the intradermal delivery route was selected to nails, or oral mucosa of the patient. Not unexpectedly (given the maximize the probability of observing an effect based on pre- small treatment area), the patient reported no significant changes clinical studies demonstrating reduced reporter gene expres- in overall quality of life during the treatment period (P = 0.16). sion after intradermal injection of specific siRNA in mice.15,16 There were no clinical signs of local or systemic toxicity at intral- Future efforts must focus on improved delivery methods for esional doses up to and including the highest TD101 dose admin- TD101, such as pharmaceutical formulations for noninvasive istered (17 mg). Similarly, no laboratory values suggested toxicity topical delivery. (Supplementary Tables S1–S4). Despite our understanding of the molecular basis of PC,17 current treatment is limited to mechanical removal of thick calluses, Discussion nonspecific topical keratolytics and oral retinoids, none of which On the basis of preclinical testing, and the dramatic and specific alleviates blistering or plantar pain satisfactorily.8 We believe the response of the patient’s treated callus to TD101, we have every callus regression in this single-patient clinical trial of siRNA in PC reason to believe, but cannot prove, that the mechanism of the is sufficiently promising to warrant additional studies of siRNA in clinical effect was through RNA interference. We were unable to this and other dominant-negative skin diseases.

Molecular Therapy 3 © The American Society of Gene & Cell Therapy siRNA Phase Ib Trial of an Inherited Skin Disorder

a TD101-Treated (right foot) weekly over 17 consecutive weeks for a total of 33 injections in each foot (Table 1). Two symmetric calluses were selected for treatment—one on Day 4 Day 115 Day 115 each foot. Test agent or vehicle-control solution was injected as indicated by the randomization list. The central region of each callus was marked and injected at the same site for each treatment; other locations on the calluses were never treated. The injection was performed using a 30- gauge needle, inserted with a single needle-stick to penetrate to the level of the superficial dermis. This depth of injection was determined by the dramatic decrease in resistance that occurred in the subepidermis. The 2 ml injections produced ~2 cm subepidermal blisters, which corresponds to the central region of response on the treated callus (Figure 3a). Beginning on day 29, in response to intense injection-related pain, the patient was premedicated with 2.5 mg diazepam and 5.0/325 mg hydrocodone/acetaminophen prior to each treatment. In addition, beginning on day 32, the patient also received bilateral posterior tibial nerve blocks with 2% preservative-free lidocaine. The treatment sites were each evaluated for adverse reactions before injections. TD101 was administered on a volume and dose-escalation schedule b Vehicle control (left foot) (Table 1). The optimal dosing schedule for unmodified siRNA in skin is not known. We based the frequency of our dosing on the stability Day 4 Day 115 Day 115 of TD101 siRNA in skin as demonstrated in our preclinical studies.15 We designed our treatment protocol to spread the injections out over a time period to optimize response and minimize the burden on the subject given the significant pain experienced by the subject during each injection. Initially, 0.1 ml of a 1.0 mg/ml solution of TD101 or vehicle alone (Dulbecco’s phosphate-buffered saline without calcium or magnesium) was administered to symmetric calluses. Six rising dose-volumes were completed without an adverse reaction to the increases: 0.1, 0.25, 0.5, 1.0, 1.5, and 2.0 ml of a 1.0 mg/ml solution of TD101 solution per injection. As the highest planned volume (2.0 ml) was well tolerated, the concentration of TD101 was then increased each week from 1 mg/ml up to a final concentration of 8.5 mg/ml. The pH of the placebo and stock study drug (10 mg/ml) was identical (7.0); furthermore, saline dilutions of up to 40- fold had no effect on pH (data not shown). The patient was followed for 3 months after the final injection, at which point the study was unblinded to both the patient and the Principal Investigator. Figure 3 Improvement of pachyonychia congenita symptoms following TD101 administration. Callus regression is seen on the right foot Study end points for safety. The maximum tolerated volume and the at the site of injection of (a) TD101 (center arrow), but not at the site of maximum tolerated dose were defined on the basis of the patient having injection of (b) vehicle on the left foot (arrows). Note that relative to the a grade 2 or higher injection site reaction (erosion, unacceptable pain, first injection, the callus on the right foot (but not on the left) developed or ulceration), or any adverse experience reported by the patient that a clearing of callus around the site of injection. The third photo in each panel is an enlargement of the site of injection from the day 115 photos. resulted in discontinuation of the study (see Supplementary Materials The callus on the right foot also shows some regression near the instep and Methods for details regarding clinical safety definitions). Safety evalu- of the foot resulting in a shortening of the total callus length. ations included assessments of adverse experiences by targeted clinical examination and clinical laboratory analyses. Clinical laboratory tests for safety were performed before first dosing (baseline), and on days 1, 46, Materials and Methods 92, 106, 114 (final injections), and 2 weeks after the final injections. These Patient enrollment and study design. After providing informed con- tests included hemogram, serum chemistry panel, antinuclear antibodies, sent, an adult participant with PC carrying a KRT6A N171K mutation C3a and Bb (complement split products), activated partial thromboplas- was enrolled in the phase Ib trial according to a protocol approved by the tin time, prothrombin time, dipstick urinalysis, and a urine pregnancy test University of Utah Institutional Review Board and the Food and Drug (see Supplementary Tables S1–S4 for study days and values of each test). Administration (IRB no. 24013 and IND no. 77504; ClinicalTrials.gov reg- No adverse clinical or laboratory events were noted during treatment or in istration no. NCT00716014; GMP (Good Manufacturing Practice) drug the 3-month follow-up period. manufactured by Agilent Technologies, Santa Clara, CA). The patient was a 39-year-old female with no history of medical problems other than her Study end points for efficacy. Measures of efficacy included weekly stan- PC symptoms. The patient served as her own control, with randomization dardized digital photography, callus and nail plate length and width mea- of TD101 or vehicle control for intradermal injection to symmetric cal- surements during each clinic visit (carbon fiber composites digital caliper; luses on opposite feet. The test agents were packaged and labeled accord- Fisher Scientific, Pittsburgh, PA), an online, time and date stamped, sub- ing to a computer-generated randomization list that assigned vehicle jective pain diary twice daily, and weekly completion of the Dermatology control or drug to the left or right foot. The decoded list of test agents was Life Quality Index12 (permission for use, AY Finlay). To evaluate poten- held in a secure place and not made available to the Principal Investigator, tial systemic effects, an assessment of the degree of follicular keratoses on patient, or other study personnel until the conclusion of the trial. the forearm and oral leukokeratosis was made at each clinic visit (better, The study evaluated safety and tolerance of multiple injections of worse, or the same). In an effort to minimize any inter-rater variability, escalating doses of TD101. Intradermal injections were given twice clinical measurements were obtained by a single investigator (S.A.L.),

4 www.moleculartherapy.org © The American Society of Gene & Cell Therapy siRNA Phase Ib Trial of an Inherited Skin Disorder

with the exception of five measurements on days 15, 50, and 114, and two Research Council (G0700314), and the British Skin Foundation. S.A.L. follow-up visits. The actual measure of the callus required the rater to visu- accepts full responsibility for the data presented in this manuscript. alize the point at which the callus ended and normal skin began. Because R.L.K., R.P.H., F.J.D.S., and W.H.I.M. have filed patents relating to short- this change is gradual, the investigator had to carefully examine the skin, interfering RNA therapy for PC. We thank Huntsman Cancer Institute and then use their best judgment to determine where the callus actually for the use of clinical facilities for this trial. This work was completed in began and ended. Salt Lake City, UT.

Statistical analysis. All statistical analysis was performed using the R sta- REFERENCES 1. Novobrantseva, TI, Akinc, A, Borodovsky, A and de Fougerolles, A (2008). Delivering tistical software version 2.6.0 (The R Foundation for Statistical Computing, silence: advancements in developing siRNA therapeutics. Curr Opin Drug Discov Devel Vienna, Austria). Scatter plots for improvement scores were augmented 11: 217–224. 2. Nguyen, T, Menocal, EM, Harborth, J and Fruehauf, JH (2008). RNAi therapeutics: with smooth curves produced by the “LOESS” function in R. The differ- an update on delivery. Curr Opin Mol Ther 10: 158–167. ence in callus length between the right and left foot was analyzed using an 3. Whitehead, KA, Langer, R and Anderson, DG (2009). Knocking down barriers: advances in siRNA delivery. Nat Rev Drug Discov 8: 129–138. autoregressive model of order one, with a time-trend term. This model uses 4. Haussecker, D (2008). The business of RNAi therapeutics. Hum Gene Ther 19: the previously occurring value as a predictor of the current value. A likeli- 451–462. hood ratio test was used to determine statistical significance of the tem- 5. Castanotto, D and Rossi, JJ (2009). The promises and pitfalls of RNA-interference- based therapeutics. Nature 457: 426–433. poral trend during the treatment period. To plot temporal trends in callus 6. DeVincenzo, J, Cehelsky, JE, Alvarez, R, Elbashir, S, Harborth, J, Toudjarska, I et al. length, we fit autoregressive models of order one to data from each foot (2008). Evaluation of the safety, tolerability and pharmacokinetics of ALN-RSV01, a novel RNAi antiviral therapeutic directed against respiratory syncytial virus (RSV). separately. The models were piecewise linear, with a change in slope at the Antiviral Res 77: 225–231. end of the dosing period, and were fit by maximum likelihood methods. 7. Kaspar, RL (2005). Challenges in developing therapies for rare diseases including pachyonychia congenita. J Investig Dermatol Symp Proc 10: 62–66. 8. Leachman, SA, Kaspar, RL, Fleckman, P, Florell, SR, Smith, FJ, McLean, WH et al. SUPPLEMENTARY MATERIAL (2005). Clinical and pathological features of pachyonychia congenita. J Investig Table S1. Hemogram Values. Dermatol Symp Proc 10: 3–17. Table S2. Serum Chemistries. 9. Smith, FJD, Kaspar, RL, Schwartz, ME, McLean, WHI and Leachman, SA (2006). Pachyonychia congenita. GeneReviews . Table S3. Urinalysis. 10. Cao, T, Longley, MA, Wang, XJ and Roop, DR (2001). An inducible mouse model for Table S4. Coagulation and Complement Parameters. epidermolysis bullosa simplex: implications for gene therapy. J Cell Biol 152: 651–656. Materials and Methods. 11. Wong, P, Domergue, R and Coulombe, PA (2005). Overcoming functional redundancy to elicit pachyonychia congenita-like nail lesions in transgenic mice. Mol Cell Biol 25: 197–205. ACKNOWLEDGMENTS 12. Hickerson, RP, Smith, FJ, Reeves, RE, Contag, CH, Leake, D, Leachman, SA et al. We are indebted to the pachyonychia congenita (PC) patient that (2008). Single-nucleotide-specific siRNA targeting in a dominant-negative skin model. J Invest Dermatol 128: 594–605. made this study possible, as well as PC Project and the membership of 13. Leachman, SA, Hickerson, RP, Hull, PR, Smith, FJ, Milstone, LM, Lane, EB et al. (2008). the International Pachyonychia Congenita Consortium for their unfail- Therapeutic siRNAs for dominant genetic skin disorders including pachyonychia ing enthusiasm and support. We thank Manuel Flores for assistance congenita. J Dermatol Sci 51: 151–157. 14. Kleinman, ME, Yamada, K, Takeda, A, Chandrasekaran, V, Nozaki, M, Baffi, JZ et al. with preparation of figures and Kristina Heintz for her nursing support. (2008). Sequence- and target-independent angiogenesis suppression by siRNA via We also acknowledge financial sponsorship of the clinical trial by PC TLR3. Nature 452: 591–597. Project and FDA (US Food and Drug Administration) OOPD (Office 15. Hickerson, RP, Vlassov, AV, Wang, Q, Leake, D, Ilves, H, Gonzalez-Gonzalez, E et al. (2008). Stability study of unmodified siRNA and relevance to clinical use. of Orphan Products Development) grant 1-R01-FD-003553-01 (to Oligonucleotides 18: 345–354. S.A.L.). This work was further supported by NIH (National Institutes of 16. Gonzalez-Gonzalez, E, Ra, H, Hickerson, RP, Wang, Q, Piyawattanametha, W, Health) grant 1R43ARO56559 (to R.L.K.), a fellowship grant from PC Mandella, MJ et al. (2009). siRNA silencing of keratinocyte-specific GFP expression in a transgenic mouse skin model. Gene Ther 16: 963–972. Project (to F.J.D.S.), and by grants to W.H.I.M. and F.J.D.S. from the 17. Smith, FJ, Liao, H, Cassidy, AJ, Stewart, A, Hamill, KJ, Wood, P et al. (2005). The Dystrophic Epidermolysis Bullosa Research Association, the UK Medical genetic basis of pachyonychia congenita. J Investig Dermatol Symp Proc 10: 21–30.

Molecular Therapy 5

editorial

Pachyonychia Congenita: Cast in Translation Journal of Investigative Dermatology (2011) 131, 995. doi:10.1038/jid.2011.52

ssembling a keratin network is no mean together, the consortium subsequently assembled feat. The process of combining acidic and a network of clinicians, clinician–scientists, and Abasic keratin monomers, stacking them scientists from a range of molecular biology, cell in an orderly fashion, and then polymerizing the biology, imaging, chemistry, physics, and pharma- bundles into rope-like intermediate filaments cological backgrounds, collectively tasked with is an intricate operation, and one that relies on improving understanding of the pathophysiology numerous cellular and subcellular processes. In of PC and driving forward new therapies that can many ways, the same organizational complexity is be delivered to patients. reflected in the International Consortium that over Central to all activities, however, has been the the past eight years has been assembled to decode PC patient. The IPCC has formulated a registry for the mysteries of the autosomal dominant keratin patients with PC and facilitated molecular screen- disorder pachyonychia congenita (PC). In this issue ing. This has led to a significant expansion in the of JID, we present a collection of eight papers that database of PC gene mutations, refinements in gen- detail the major clinical, diagnostic, mechanis- otype–phenotype correlation, a new classification tic, and therapeutic advances that have emerged for PC, and a greater understanding of the impact of in recent years that were presented at the annual PC on patient quality-of-life issues, such as severity meeting of the International PC Consortium (IPCC) of plantar pain. But what perhaps was needed most in May 2010. was a clear strategy of how to develop therapies. The earliest clinical descriptions of PC appeared Modeling of PC established that mutant allele- more than 100 years ago, and the first caus- specific silencing using RNAi approaches might ative mutations were reported in the mid 1990s. be a suitable method for treating PC, and in 2010 Subsequently, PC has been shown to involve the first human trial of siRNA for an inherited skin heterozygous mutations in the KRT6A, KRT6B, disease was carried out with clear clinical benefits KRT16, and KRT17 genes that encode keratins in the treated area. Evidence was also established K6a, K6b, K16, and K17, respectively. Of course, for some functional redundancy among certain a similar tale of careful clinical definition followed keratins, indicating that a more generic approach by rigorous molecular dissection has emerged for in completely silencing a defective keratin was several other Mendelian skin diseases. But where also likely to be a rational therapeutic endeavor the story of PC differs from that of most other gen- in some other PC cases. Overhanging much of the odermatoses is in what happened next. molecular targeting, however, has been the major Establishing the molecular basis of disease pro- obstacle of effective delivery to target keratino- vides a platform for creating in vitro and in vivo cytes. This is a current priority for the IPCC and a disease models, as well as translational benefits in focus for its assembled crew of “delivery” experts. terms of improved diagnostics and genetic coun- As a model for how to perform translational seling, but often pushing the translational envelope research, the PC work has established a new fails to overcome many of the obstacles blocking paradigm for progress. Germane to all advances, major benefits for patients, particularly toward the however, has been the close engagement with development of novel therapies. Thankfully for PC the patient advocacy group PC Project, as well as investigators and patients, however, substantial PC patients themselves, and the multidisciplinary progress has been made, and key to this success group of clinicians and scientists. Assembling an has been the assembly of a multifunctional profes- impressive cast of capable performers has been a sional cast, each with subspecialist skills sufficient considerable achievement, and now the stage is to overcome therapeutic inertia or stalling. set. An expectant audience of patients and inves- Fundamental to the translational push has tigators eagerly awaits the next act. been the activity of the patient advocacy group PC Project (http://www.pachyonychia.org), which John A. McGrath helped establish the IPCC in 2004. Working Deputy Editor

REVIEW

The Phenotypic and Molecular Genetic Features of Pachyonychia Congenita W.H. Irwin McLean1, C. David Hansen2, Mark J. Eliason2 and Frances J.D. Smith1

Pachyonychia congenita (PC) is an autosomal domi- Lewandowski, 1906; Jackson and Lawler, 1951) but it was nant genodermatosis caused by heterozygous not until the early 1990s, with the emergence of molecular mutations in any one of the genes encoding the genetics technology, that the causative gene in a large differentiation-specific keratins K6a, K6b, K16, or K17. Scottish PC family was mapped to one of the keratin gene The main clinical features of the condition include clusters (Munro et al., 1994). Shortly thereafter, the causative painful and highly debilitating plantar keratoderma, mutations were identified in several PC patients in the KRT6A, hypertrophic nail dystrophy, oral leukokeratosis, and a KRT6B, KRT16,andKRT17 genes, encoding the keratin variety of epidermal cysts. Although the condition proteins K6a, K6b, K16, and K17, respectively (Bowden has previously been subdivided into PC-1 and PC-2 et al., 1995; McLean et al., 1995; Smith et al., 1998). subtypes, the phenotypic characterization of 1,000 Keratins are the intermediate filament proteins specifically expressed by epithelial cells, in which they form a dense mutation-verified PC patients enrolled in the Interna- cytoplasmic network (Irvine and McLean, 1999; Omary et al., tional PC Research Registry, coordinated by the 2004). The primary function of the keratin cytoskeleton is to patient advocacy group PC Project, shows that there impart mechanical strength and resilience to epithelial cells is considerable overlap between these subtypes. and tissues. Disruption of this cytoskeletal system due to a Thus, a new genotypic nomenclature is proposed, in genetic mutation in a keratin gene leads to extreme fragility of which PC-6a represents a patient carrying a mutation the epithelial cells and tissues in which the mutated keratin is in the K6a gene, etc. Although a rare disorder, PC expressed. Similar to several other keratin disorders, the represents a good model for therapy development, vast majority of causative mutations in the PC-related keratins and international efforts are ongoing to develop and are heterozygous missense mutations or small insertion/ deliver siRNA, gene, correction, small molecule, and deletion mutations that disrupt cytoskeletal function via other strategies to treat this painful, disabling dominant-negative interference and lead to epithelial cell skin condition. The special relationship between PC fragility (McLean et al., 2005). In PC, this is manifest as Project and the PC research community has greatly cytolysis and hyperkeratosis in the subset of differentiated accelerated the development pathway from gene epithelial tissues in which K6a, K6b, K16, and K17 are identification to clinical trials in only a few years and predominantly expressed (Lane, 1993), specifically the represents a paradigm of hope for other orphan palmoplantar epidermis, nail bed, mucosae, and the pilose- diseases. baceous unit. Thus, the cardinal phenotypic features of PC are palmoplantar (predominantly plantar) keratoderma; Journal of Investigative Dermatology advance online publication, 24 March 2011; doi:10.1038/jid.2011.59 hypertrophic nail dystrophy; oral leukokeratosis; and a variety of cysts arising from hyperkeratosis of pilosebaceous apparatus (Figure 1). INTRODUCTION Pachyonychia congenita (PC) is an uncommon autosomal A MOLECULAR CLASSIFICATION FOR PC SUBTYPES dominant disorder of keratinization caused by mutations in In 2003, a patient advocacy group—Pachyonychia Congenita any one of a number of keratin genes that are expressed Project—was established to support those affected by PC in differentiated epithelial tissues. The condition was first and to both encourage and fund research into a cure for the described in the early twentieth century (Jadassohn and condition (www.pachyonychia.org). To achieve this goal, the International PC Consortium (IPCC) was founded in early 1Division of Molecular Medicine, University of Dundee, Dundee, UK and 2004. This is a group of clinicians and scientists actively 2Department of Dermatology, University of Utah, Salt Lake City, Utah, USA researching the causes of PC and importantly, the development Correspondence: W.H. Irwin McLean, Division of Molecular Medicine, of new treatments for PC. The IPCC has met annually since Medical Sciences Institute, University of Dundee, Dundee DD1 5EH, UK. 2004 and its membership is listed at www.pachyonychia.org. E-mail: [email protected] An important part of the ongoing PC research program is Abbreviations: IPCC, International Pachyonychia Congenita Consortium; the International PC Research Registry (IPCRR), in which IPCRR, International Pachyonychia Congenita Research Registry; K, keratin protein; KRT, keratin gene; PC, pachyonychia congenita detailed phenotypic data are collected from patients and Received 26 August 2010; revised 2 February 2011; accepted 11 February linked to genetic data. At the time of writing, close to 1000 2011 PC patients have been identified by the PC Project. This has

& 2011 The Society for Investigative Dermatology www.jidonline.org 1 W.H. Irwin McLean et al. Clinical and Genetic Features of PC

Figure 1. The cardinal clinical characteristics of PC. (a) Focal plantar keratoderma, with recurrent blistering underneath the callus, is the main source of pain and disability in PC. This individual is a heterozygous carrier of the K16 mutation L132P (p.Leu132Pro). (b) Typical hypertrophic nail dystrophy that gives the condition its name. The nail changes are quite variable in PC families, even among people with the same mutation. In some cases, the fingernails are spared. This patient carries the K17 mutation L95Q (p.Leu95Gln). (c) Oral leukokeratosis is a common feature of PC, readily seen here as lingual leukokeratosis in a patient carrying the most common PC mutation, N172del (p.Asn172del) in K6a. (d) PC patients suffer from a variety of epidermal cysts, including follicular keratoses and pilosebaceous cysts that can resemble steatocysts or epidermoid cysts. The latter are more abundant in patients with K17 mutations, as shown here in a patient carrying the K17 mutation N92D (p.Asn92Asp). PC, pachyonychia congenita.

led to the largest collection of linked clinical and genetic problematic symptom reported by PC patients is focal plantar information yet assembled for a rare keratin disorder. keratoderma that is associated with severe pain. The plantar Historically, PC has been split into two subtypes (PC-1, or pain in PC is often highly debilitating and has considerable Jadassohn–Lewandowski subtype; and PC-2, or the Jackson–- negative effect on quality of life. The reason for the pain is not Lawler subtype) on the basis of subtle differences in fully understood but is thought to be related to blister phenotype, primarily the presence or absence of pilosebac- formation deep underneath the thick callus that develops eous cysts (Jadassohn and Lewandowski, 1906; Jackson over the pressure points of the plantar surface (see Figure 1a). and Lawler, 1951). At present, with analysis of hundreds of Plantar blistering, together with accompanying pain, is a patients in the IPCRR, limitations in the older classification, common feature of PC that is under-reported in the literature which was based on only a handful of non-genotyped cases, (Eliason et al., unpublished data). have become clear. In particular, many PC patients, Nail dystrophy (Figure 1b), which can occur from a very regardless of genotype, have some form of epidermal cysts early age, is variable in severity and in many cases not all 20 (see Wilson et al., 2011). Therefore, on the basis of the more nails are affected. Toenails are more commonly affected than comprehensive IPCRR data, a more rational and useful fingernails, which could be because of greater trauma exerted classification based on the mutated gene was proposed at from shoes. Another feature of PC is oral leukokeratosis the 2010 IPCC Symposium and has been adopted throughout (Figure 1c). This is often one of the first signs of PC in babies the research papers in this issue of the JID.Thenew and may lead to difficulty in feeding and is often mistaken for classification is PC-6a, PC-6b, PC-16, and PC-17, for a patient candidiasis in infants. Follicular keratoses are present in with a mutation in the gene encoding K6a and others proteins many cases of PC. Some individuals also develop cysts in the (the complete data set underlying this new nomenclature form of steatocysts (steatocystomas) and/or pilosebaceous will be published elsewhere; Eliason et al., unpublished data). cysts (Figure 1d). This feature is particularly associated with In cases in whom PC is suspected but no mutation has been patients with a K17 mutation (see Wilson et al., this issue), in found (or not looked for), the term PC-U (for unknown) will whom sometimes it is necessary to remove cysts surgically. be used. The severity of the clinical features of PC can vary quite widely both among and within families. This may partly be The predominant symptom in PC is plantar pain because of individual lifestyle and care of PC and could also Although hypertrophic nail dystrophy is the phenotypic be because of the specific type of mutation, as well as other feature that gave rise to the name of the condition, the most genetic and/or environmental factors.

2 Journal of Investigative Dermatology W.H. Irwin McLean et al. Clinical and Genetic Features of PC

Rapid therapy development in PC ACKNOWLEDGMENTS Although it is a rare condition, PC is at the forefront of We thank Pachyonychia Congenita Project for their continued enthusiasm, genetic therapy development in the dermatology field. constant encouragement, and financial support of their research. In particular, the dominant-negative genetic mechanism in PC contributes to therapeutic strategies based on RNA REFERENCES interference (RNAi), especially in the form of short interfering Bowden PE, Haley JL, Kansky A et al. (1995) Mutation of a type II keratin gene RNA (siRNA). It has been demonstrated that mutant keratin (K6a) in pachyonychia congenita. Nat Genet 10:363–5 alleles differing from wild type by a single-nucleotide point Hickerson RP, Smith FJD, Reeves RE et al. (2008) Single-nucleotide-specific mutation can be potently and specifically silenced siRNA targeting in a dominant-negative skin model. J Invest Dermatol by carefully designed siRNA (Hickerson et al., 2008). This 128:594–605 mutation-specific siRNA therapy approach has been pro- Irvine AD, McLean WHI (1999) Human keratin diseases: the increasing gressed into the recently reported small-scale human clinical spectrum of disease and subtlety of the phenotype-genotype correlation. Br J Dermatol 140:815–28 trial, in which efficacy was demonstrated (Leachman et al., Jackson ADM, Lawler SD (1951) Pachyonychia congenita: a report 2010). This was the first time that siRNA had been used to of six cases in one family with a note on linkage data. Ann Eugen treat a human skin disorder. The keratins involved in PC also 16:142–6 show considerable functional redundancy, in particular when Jadassohn J, Lewandowski F (1906) Pachyonychia congenita: keratosis K6 is involved. Humans have three copies of a nearly disseminata circumscripta (follicularis). Tylomata. Leukokeratosis lin- identical KRT6 gene, encoding the K6a, K6b, and K6c guae. vol. 1. Berlin: Urban and Schwarzenberg, 29 proteins. Mouse knockout experiments strongly suggest that Kaspar RL, Leachman SA, McLean WHI, Schwartz ME (2011) Toward a Treatment for Pachyonychia Congenita: Report on the 7th Annual loss of one of these keratins may be tolerated (Wong et al., International Pachyonychia Congenita Consortium Meeting. J Investig 2000; Wojcik et al., 2001) and therefore an alternative Dermatol 131:1011–4 therapeutic approach would be to completely silence the Lane EB (1993) Keratins. In Connective Tissue and its Heritable Disorders. defective keratin, regardless of mutation. To this end, highly Molecular, Genetic and Medical Aspects. Royce PM, Steinmann B (eds). potent gene-specific siRNA has been developed for PC (Smith New York: Wiley-Liss Inc, 237–47 et al., 2008). The major technical hurdle yet to overcome in Leachman SA, Hickerson RP, Schwartz ME et al. (2010) First-in-human mutation-targeted siRNA phase Ib trial of an inherited skin disorder. both these therapeutic approaches is the development of a Mol Ther 18:442–6 safe, effective, and patient-friendly method for routine McLean WHI, Rugg EL, Lunny DP et al. (1995) Keratin 16 and keratin delivery of siRNA into the epidermis. This is currently a 17 mutations cause pachyonychia congenita. Nat Genet 9:273–8 major goal of the IPCC research groups, in addition to McLean WHI, Smith FJD, Cassidy AJ (2005) Insights into genotype- development of alternative therapies that include gene phenotype correlation in pachyonychia congenita from the human correction methodology, small molecule therapy, and other intermediate filament mutation database. J Investig Dermatol Symp Proc 10:31–6 strategies (Kaspar et al., this issue). Munro CS, Carter S, Bryce S et al. (1994) A gene for pachyonychia congenita is closely linked to the keratin gene cluster on 17q12-q21. J Med Genet 31:675–8 CONCLUSION Omary MB, Coulombe PA, McLean WHI (2004) Intermediate Over the past 7 years, the PC research field represents a great filament proteins and their associated diseases. N Engl J Med 351: example of how a small group of highly motivated patients and 2087–100 their families, together with a group of interested clinicians and Smith FJD, Hickerson RP, Sayers JM et al. (2008) Development of scientists, can rapidly progress research from knowing only the therapeutic siRNAs for pachyonychia congenita. J Invest Dermatol 128: identity of a gene defect to having new therapies that show 50–8 efficacy in cells, in animal models, and in patients. Hopefully, Smith FJD, Jonkman MF, van Goor H et al. (1998) A mutation in human keratin K6b produces a phenocopy of the K17 disorder pachyonychia this sustained, highly focused effort will shortly lead to a congenita type 2. Hum Molec Genet 7:1143–8 successful and widely applicable treatment for PC and the Wilson NJ, Leachman SA, Hansen CD et al. J Invest Dermatol; e-pub ahead of lessons learned along the way can be translated to other genetic print 17 February 2011 skin disorders. Wojcik SM, Longley MA, Roop DR (2001) Discovery of a novel murine keratin 6 (K6) isoform explains the absence of hair and nail defects in mice deficient for K6a and K6b. J Cell Biol 154:619–30 CONFLICT OF INTEREST Wong P, Colucci-Guyon E, Takahashi K et al. (2000) Introducing a null Dr McLean and Dr Smith filed a patent on therapeutic siRNA for PC. The mutation in the mouse K6alpha and K6beta genes reveals their essential other authors state no conflict of interest. structural role in the oral mucosa. J Cell Biol 150:921–8

www.jidonline.org 3

FEATURE ARTICLE Pachyonychia Congenita Project A Partnership of Patient and Medical Professional Mary E. Schwartz, Gail M. Zimmerman, Frances Smith, Eli Sprecher

ABSTRACT: A rare disease like pachyonychia congenita proud to know how far we’ve come. No longer (PC) poses barriers to the patient, medical professional, are we sitting on fingers at the dinner table, and scientist. The patient has challenges connecting to hiding our PC as she didIinstead we are information, the medical professional has challenges sharing with anyone information about PC, connecting to patient experience, and the scientist has proud that we have an organization that challenges connecting to a sufficient number of pa- stands by us. And, most importantly, we are tients to do meaningful research. Recent collaboration not alone with PC. It may be rareIbut we between these groups has transformed our under- have met friends with it thanks to last year’s standing of PC and its symptoms and method of diag- PC Patient Support Meeting. We are not nosis. PC Project is at the center of this collaboration alone!’’ (Jessica, pachyonychia congenita and is providing new insights for the dermatologist and [PC] patient). dermatology nurse, enabling better diagnosis of PC and counseling of a PC patient. The PC patient, medi- hese words of joy reflect the power that can cal professional, and scientist have an international come from shared experienceVknowing you advocate in PC Project, a patient-led, nonprofit project are part of a larger community. This joy of committed to connecting all these communities to the knowing can be especially empowering when tools they need to improve the lives of those living with PC. living with a rare disease like PC. Key words: Blisters, Cysts, Keratoderma, Nail dystrophy, PC is an ultra-rare skin disorder that affects 2,000Y10,000 Pachyonychia, Partnership Tpeople worldwide. The rarity of PC means that a person with PC usually never meets anyone else with PC other ‘‘I was one proud mama! I watched my than affected family members. It means that researchers son hand PC brochures out to people (at are challenged to find a sufficient number of PC patients our PC Awareness Day Event). When they to do meaningful research. It means that medical pro- asked questions, he answered them. If my fessionals do not have easy access to broad patient expe- grandmother was still alive she would be so rience on which to base diagnosis or to learn if a treatment really works. Mary E. Schwartz, AA, Pachyonychia Congenita Project, Salt In 2004, PC Project [a patient-led 50l(c)(3) charity based Lake City, Utah. in Salt Lake City, Utah] set out to eliminate these barriers Gail M. Zimmerman, BA, Pachyonychia Congenita Project, Salt of isolation by creating an international collaborative net- Lake City, Utah. work of patients, medical professionals, and scientists to Frances Smith, PhD, Division of Molecular Medicine, University be a catalyst to find effective treatments for the disease. of Dundee, Dundee, United Kingdom. Eli Sprecher, MD, PhD, Department of Dermatology, Tel Aviv Using high-technology tools, including custom software Sourasky Medical Center, Tel Aviv, Israel, and Sackler Faculty of for the PC registry and database, an interactive Web site Medicine, Tel Aviv University, Ramat Aviv, Israel. (www.pachyonychia.org), Webinars, and Web meetings Conflicts of interest and sources of funding: There are no conflicts (as well as E-mail, conference calls, and VoIP services or of interest. PC Project is a 501(c)(3) U.S. public charity. Funding Skype), PC Project has connected researchers interested sources are from donations as well as grants from the National in keratin disorders, multidisciplinary specialists, and der- Institutes of Health. matologists to conduct basic and clinical research. This Correspondence concerning this article should be addressed to Mary E. Schwartz, AA, Pachyonychia Congenita Project, 2386 E. group of specialists is known as the International PC Heritage Way, Suite B, Salt Lake City, UT 84109. Consortium (IPCC). PC Project has also identified and E-mail: [email protected] recruited patient volunteers to provide personal histories DOI: 10.1097/JDN.0b013e31827d9ed5 and has in turn provided physician consultations and

42 Journal of the Dermatology Nurses’ Association

Copyright © 2013 Dermatology Nurses' Association. Unauthorized reproduction of this article is prohibited. genetic testing for these patients through the International refers to a spectrum of symptoms, which is determined by PC Research Registry (IPCRR). PC Project continues to both the location and the nature of the causative mutation seek dermatologists and dermatology nurses to be part- as well as by yet unknown additional genetic and envi- ners in this international community. ronmental factors (which explain the fact that individuals Through this collaborative effort, researchers have made carrying the very same mutation can display divergent clin- several discoveries that have transformed the understanding ical features). From detailed questionnaires gathered from of PC and treatment for PC. These new insights have sig- nearly 500 patients participating in the IPCRR, those with nificance for the dermatologist and dermatology nurse in genetically confirmed PC are consistently found to have a diagnosing and counseling a PC patient. triad of features including nail dystrophy, palmar/plantar keratoderma (e.g., thickening of the skin), and pain. PC: WHAT WE HAVE LEARNED PC shares symptoms with a number of disorders like The disease is characterized by accumulation of keratin epidermolysis bullosa simplex and some connexin disor- in the skin and nails, which manifests as calluses, thick- ders (e.g., Clouston syndrome). In PC, the unrelenting ened nails, and cysts (see Figures 1Y3). Under the calluses pain from blisters under the thick calluses on the soles of are extremely painful blisters on the soles and sometimes the feet is the principal life-altering feature for most with on the palms. Pain is one of the most consistent features of this disease. The callus usually begins when the child with PC and a key in differential diagnosis (Eliason, Leahman, PC first begins to walk, and the pain is usually constant by Feng, Schwartz, & Hansen, 2012). PC is caused by a sin- age of 10 years. It requires regular trimming of the callus gle mutation in one of at least five keratin genes, KRT6A, and activity planning to limit time on one’s feet. Many KRT6B, KRT6C, KRT16, or KRT17 (Akasaka et al., must use canes, crutches, or wheelchairs or must crawl on 2011; Wilson et al., 2010, 2011). As PC is an autosomal- their knees to manage the pain. In some forms of PC, the dominant disorder, there is a 50% chance of passing on prevalent cysts are the most painful feature. the mutation with each conception. However, more than Documenting the variability of the symptoms across 45% of PC cases appear spontaneously where there is no mutations and genes has led to a new classification system family history of the disease. of PC to help clinicians improve their diagnostic and prog- The condition was first described in the early 20th cen- nostic accuracy and improve therapeutic development. tury (Jadassohn & Lewandowski, 1906) as ‘‘pachyo- Rather than refer to PC-l and PC-2, which inaccurately nychia,’’ which means ‘‘thickened nails.’’ However, PC groups PC patients, the new nomenclature, a diagnosis of

FIGURE 1. Focal plantar keratoderma, with painful blistering underneath the callus, is the main source of pain and disability in PC. Palmar keratoderma may also be found in patients with PC. Abbreviation: PC = pachyonychia congenita.

VOLUME 5 | NUMBER 1 | JANUARY/FEBRUARY 2013 43

Copyright © 2013 Dermatology Nurses' Association. Unauthorized reproduction of this article is prohibited. FIGURE 2. Nail changes in PC are quite varied across PC patients even among family members or those with the same mutations. Abbreviation: PC = pachyonychia congenita.

PC-K6a, PC-K6b, PC-K6c, PC-K16, and PC-K17, corre- contact us directly or refer the patient to PC Project. Indi- sponds to mutations in the KRT6A, KRT6B, KRT6C, vidual consultations for patients, physicians, and derma- KRT16, and KRT17 genes, respectively, and PC-U indi- tology nurses treating PC patients may be scheduled with cates those with an unconfirmed genetic mutation (McLean, physicians serving on the PC Project Medical and Scien- Hansen, Eliason, & Smith, 2011). tific Advisory Board. Once the diagnosis of PC is confirmed, the PC patient’s de-identified data will be made available to researchers THE ROLE OF THE DERMATOLOGY NURSE around the world (while preserving patient privacy and Although there is no effective treatment for PC at this anonymity), and the patient will be eligible to participate time, the dermatology nurse can be a very helpful partner in future studies and clinical trials. to the PC patient by helping with diagnosis, care techniques, and provision of accurate information. Those with Diagnosis PC may require medical assistance when they experience an infection of the nails or cysts. A good relationship with It is not uncommon for a person to reach his or her 40s or a dermatology nurse can be extremely beneficial to pa- 50s before getting a correct diagnosis. Data from patients tients with PC. participating in the IPCRR have now revealed that plantar Also, the dermatology nurse can effectively assist the pain is a key diagnostic clue. In a recent study of 254 PC patient in participating in the IPCRR. Through the IPCRR, patients, plantar pain was reported by 225 of 254 (89%) the patient can obtain genetic testing to determine whether surveyed patients. Only three patients older than 10 years the patient has a mutation in one of the genes associated old did not report plantar pain. The researchers concluded with PC. Details regarding the gene and the specific mu- that plantar pain is the most important feature of PC af- tation are provided with the test results, and genetic counsel- fecting quality of life. The pain is related to the underlying ing is available. If the patient is found not to have PC, blisters, and the thickness or extent of the visible callus genetic testing is conducted for a number of other similar may not reflect the degree of pain a patient experiences conditions with the overall goal to assist the patient and (Eliason et al., 2012). medical providers with a definitive answer regarding the condition. All PC Project services are provided at no cost Main Features of PC to the patient and referring specialist. Because of the IPCRR, the clinician who only rarely en- The IPCC experts welcome the opportunity to support counters these disorders now has better guidance and a the nurse and/or doctor in diagnosis, treatment options, more robust framework to make a clinical diagnosis (Irvine, and care. The dermatologist or dermatology nurse may 2012, p. 1758).

FIGURE 3. PC patients experience a variety of epidermal cysts. Those with PC-K17 have the greatest abundance of steatocystomas. Abbreviation: PC = pachyonychia congenita.

44 Journal of the Dermatology Nurses’ Association

Copyright © 2013 Dermatology Nurses' Association. Unauthorized reproduction of this article is prohibited. Nail dystrophy (or thickening) often does not affect may benefit from oral retinoids, an empirical trial with all 10 fingernails or all 10 toenails. variable dosing is being evaluated as well as a topical ret- Only toenails may be affected with fingernails re- inoid application for PC. maining normal. Patients typically manage their own symptoms through Calluses on the soles of feet have underlying hidden careful mechanical trimming of the calluses to ensure a blisters and are usually extremely painful with plantar ‘‘not too thick and not too thin’’ result. Aggressive debride- pain being the number one concern for PC patients. ment may increase pain. Frequent filing and trimming of Cysts are the second most common cause of pain in the nails is also necessary. Patients who experience PC with PC, and pain may also be from infected nails or fol- numerous cysts find that these must be drained or the cysts licular hyperkeratosis. may need to be surgically removed because of pain or Both steatocystoma and pilosebaceous cysts are found infection. in nearly all types of PC, although those with PC-K17 The results of a formal survey of patient experience with have a greater number of steatocystomas. treatment and self-care will also be published soon. The Nail infections may be initiated by nail trauma or ex- survey collected detailed responses on treatment and care cessive trimming. used by PC patients for specific conditions including keratoderma; cysts; follicular hyperkeratosis; and fingernails, toenails, and nail infections. The results will be published Other Features of PC shortly and will provide a good reference for both the Follicular hyperkeratosis (most prominent in chil- dermatologist and dermatology nurse. dren and young teens) Patients may need one or more medical services on a Oral leukokeratosis (often misdiagnosed as thrush or regular basis including quick access to prescriptions for leukoplakia) antibiotics when infections arise, removal of cysts, pain Prenatal or natal teeth (usually PC-K17) medication, or other care related to the varied symptoms Acute pain related to ‘‘first bite syndrome’’ (often mis- of PC. The dermatology nurse/dermatologist may need to diagnosed as ‘‘ear’’ pain) help the PC patient find treatment for one or more of these Laryngeal problems: thickening or nodules on the issues. vocal chords causing hoarseness or, sometimes, difficulty in breathing (Treatment of the larynx may cause increased overgrowth but may sometimes be neces- PC PROJECT: A RESOURCE FOR THE PC PATIENT sary to avoid obstruction.) AND MEDICAL PROFESSIONAL Major benefits for PC patients and medical professionals The rarity of the disease and its overlapping symptoms are available through the PC Project, the IPCRR, and with other skin disorders make diagnosis a challenge. IPCC members. Participation in the IPCRR provides ac- Commonly, people with PC may be misdiagnosed with cess to clinical trials, information on care techniques, publi- onychomyciosis or fungal infection of the nails or with cations from dermatology journals, annual patient support other causes for palmar/plantar keratoderma. At the same meetings, and Webinars. Through the IPCC, medical pro- time, people with only thickened or dystrophic nails may fessionals have a team of support and information for their be told that they have PC. Furthermore, previously mis- patients’ care. attributed features such as deafness, mental retardation, Through the Web site (www.pachyonychia.org), both diabetes, bony abnormalities, early menarche, corneal le- the patient and medical professional can stay informed sions, and cataracts can be safely excluded from the canon about opportunities to participate in educational meet- of PC manifestations (Irvine, 2012, p. 1758). ings, Webinars, and clinical trials. The Web site features Genetic testing can confirm the clinical diagnosis of PC the latest research news, extensive images, a complete bib- and is the only way to verify that the patient’s condition is liography of scientific articles, patient education brochures, in fact PC and not a related disorder. a quarterly scientific newsletter, and a monthly news brief as well as an annual report. PC Project also maintains a Treatment Facebook page through which PC patients in the IPCRR Review of the experience of PC patients in the IPCRR has can connect with each other. shown that, although traditional therapeutics such as Patient success with various care techniques does vary. urea, salicylic acid, or oral retinoids may soften or reduce To help PC patients find what may work for them, PC calluses, most patients abandon those therapies for lack of Project has catalogued patient reports into a PC Wiki sec- sufficient benefit (Eliason et al., 2012). When oral reti- tion available on the Web site. Techniques used to manage noids were recently assessed, findings show that, for most the symptoms of PC are shown on a ‘‘Caring For PC’’ patients, there was no benefit and, instead, increased pain DVD, which is available upon request. The DVD and all or the adverse side effects outweighed the benefits (Gruber printed materials are provided at no charge to patients and et al., 2011). Because we cannot predict which patients medical professionals.

VOLUME 5 | NUMBER 1 | JANUARY/FEBRUARY 2013 45

Copyright © 2013 Dermatology Nurses' Association. Unauthorized reproduction of this article is prohibited. The annual PC Patient Support Meeting (in Europe or RAISING RARE DISEASE AWARENESS the United States) is an especially important way a PC pa- As PC is ultra-rare, the public and medical community are tient can meet, connect, and share with others who have generally not aware of PC or the challenges it can present. PC. To help every PC patient who wants to attend, PC This lack of awareness contributes to the isolation of many Project offers scholarships to patients in the IPCRR to PC patients. The empathy and understanding of the general attend the meeting. public will create a more supportive environment for those To learn directly from physician experts about PC, in- experiencing rare diseases like PC and generate more public terested dermatology nurses can sign up for one of the support of rare disease research. quarterly educational Webinar sessions sponsored by PC Raising funds for rare disease research and patient sup- Project. There are also educational brochures specifically port has been challenging in terms of recruiting volunteers for medical professionals. In addition, for patients in the and supporters. However, because of the profound impact IPCRR, medical experts are available to provide counsel- this disease can have on the life of both children and adults, ing with a physician or nurse to individual patients. parents, spouses, and family members are highly moti- We recognize that most medical professionals will rarely vated to help improve their future. PC Project launched its see a PC patient, so PC Project has set up a resource for first annual PC Awareness Day in 2012 to begin connect- those medical professionals wanting to confirm diagnosis ing and empowering the PC patient community in its or get the latest advice on management of the disease or support of public awareness. This first international effort clinical trials that may be ongoing. resulted in dozens of local events in 10 countries and reached thousands around the world through print, TV, and radio media. PC Project just completed a 5- to 6- WORKING TO CHANGE THE FUTURE: MAINTAINING minute public awareness video on PC that can be found RESEARCH MOMENTUM on YouTube. PC Awareness Day provided a new source of PC Project is collaborating on a number of research strat- funds as well as an opportunity to increase awareness of egies to develop an effective PC treatment. One of these PC. All 2012Y2013 donations are matched $2-for-$1 by research strategies is gene silencing. By inactivating the an anonymous individual sponsor, which triples each dollar mutant keratin gene, researchers hope to eliminate PC received. symptoms. PC Project, through its biotech partner Trans- PC Project participates with other nonprofit groups, Derm, has completed a Phase 1b, FDA-approved clinical which specialize in areas important to PC research and trial of a gene silencing strategy using siRNA therapy patient support. These include Genetic Alliance (focuses (Leachman et al., 2010). A second clinical trial of an im- on genetic disorders), National Organization for Rare proved siRNA developed by TransDerm will begin in Disorders (focuses on rare disorders), and the Coalition 2013 (Hickerson et al., 2011). This second trial will in- of Skin Diseases (focuses on skin disease research and volve use of dissolvable microneedles developed to de- patient advocacy). The American Academy of Dermatol- liver drugs in a more patient-friendly manner (Gonzalez ogy and the Society of Investigative Dermatology are et al., 2010). Two additional clinical studies (oral statins important partners in the effort to raise awareness and and topical rapamycin) will also enroll patients in 2013. provide support for skin disease research. The best part PC Project provides fellowship and grant funding to a of PC Project is the partnershipsVhelping people con- variety of universities for the support of PC research. Since nect with each other and to connect the patients and 2004, when the IPCC was formed with 23 physicians and scientists. PC Project was started to help one person, but researchers, PC Project has recruited 150 members of the now, it serves and unites more than 1,000 people with PC consortium, each with a pledge to collaborate to improve in over 50 countries. People with PC need to know that, treatment for PC. A number of ground-breaking basic and by sharing information and working with others in the PC clinical research reports have been published in the last community, they can advance research and change the couple of years by members of the IPCC in leading der- future. The growing number of patients, physicians, and matology journals. These reports are accessible to every- scientists in our network of collaborators will move us one through a searchable bibliography on the PC Project’s more rapidly to an effective, patient-friendly treatment. h Web site or can be requested on a CD or in booklet format. Topics of articles, which will be published shortly based on the results of current studies, include PC best practices, REFERENCES life history of PC, pathology of PC nails, pathology of PC Akasaka, E., Nakano, H., Nakano, A., Toyomaki, Y., Takiyoshi, N., Rokunohe, D., & Sawamura, D. (2011). Diffuse and focal palmo- cysts, and over 50 additional case studies including novel plantar keratoderma can be caused by a keratin 6c mutation. British mutations as well as a revised PC summary article. The Journal of Dermatology, 165(6), 1290Y1292. doi: 10.1111/j.1365-2133 .2011.10552.x. National Institutes of Health Office of Rare Disease Re- Eliason, M. J., Leachman, S. A., Feng, B. J., Schwartz, M. E., & Hansen, search and National Institutes of Health/National Institute C. D. (2012). A review of the clinical phenotype of 254 patients with genetically confirmed pachyonychia congenita. Journal of the American of Arthritis and Musculoskeletal and Skin have been very Academy of Dermatology, 67, 680Y686. doi: 10.1016/j.jaad.2011 helpful to PC Project in our research and outreach efforts. .12.009.

46 Journal of the Dermatology Nurses’ Association

Copyright © 2013 Dermatology Nurses' Association. Unauthorized reproduction of this article is prohibited. Gonzalez-Gonzalez, E., Speaker, T. J., Hickerson, R. P., Spitler, R., Flores, circumscripta (follicularis). In Tylomata. Lukokeratosis linguage (Vol. 1, M. A., Leake, D., & Kaspar, R. L. (2010). Silencing of reporter gene p. 29). Berlin, Germany: Urban and Schwarzenberg. expression in skin using siRNAs and expression of plasmid DNA de- Leachman, S. A., Hickerson, R. P., Schwartz, M. E., Bullough, E. E., livered by a soluble protrusion array device (PAD). Molecular Therapy, Hutcherson, S. L., Boucher, K. M., & Kaspar, R. L. (2010). First-in- Y 18(9), 1667 1674. doi: 10.1038/mt.2010.126. human mutation-targeted siRNA phase 1b trial of an inherited skin dis- Gruber, R., Edlinger, M., Kaspar, R. L., Hansen, C. D., Leachman, S., order. Molecular Therapy, 18(2), 442Y446. doi: 10.1038/mt.2009.273. Milstone, L. M., & Schmuth, M. (2011). An appraisal of oral retinoids McLean, W. H., Hansen, C. D., Eliason, M. J., & Smith, F. J. (2011). The in the treatment of pachyonychia congenita. Journal of the American phenotypic and molecular genetic features of pachyonychia congenita. Y Academy of Dermatology, 66,e193 199. doi: 10.1016/j.jaad.2011.02.003. Journal of Investigative Dermatology, 131(5), 1015Y1017. doi: 10.1038/ Hickerson, R. P., Flores, M. A., Leake, D., Lara, M. F., Contag, C. H., jid.2011.59. Leachman, S. A., & Kaspar, R. L. (2011). Use of self-delivery siRNAs Wilson, N. J., Leachman, S. A., Hansen, C. D., McMullan, A. C., to inhibit gene expression in an organotypic pachyonychia congenita Y Milstone, L. M., Schwartz, M. E., & Smith, F. J. (2011). A large mu- model. Journal of Investigative Dermatology, 131(5), 1037 1044. doi: tational study in pachyonychia congenita. Journal of Investigative 10.1038/jid.2010.426. Dermatology, 131(5), 1018Y1024. doi: 10.1038/jid.2011.20. Irvine, A. D. (2012). Double trouble: Homozygous dominant mutations Wilson, N. J., Messenger, A. G., Leachman, S. A., O’Toole, E. A., Lane, E. and hair loss in pachyonychia congenita. Journal of Investigative Der- B., McLean, W. H., & Smith, F. J. (2010). Keratin K6c mutations cause Y matology, 132(7), 1757 1759. doi: 10.1038/jid.2012.121. focal palmoplantar keratoderma. Journal of Investigative Dermatology, Jadassohn, J. L. F. (1906). Pachyonychia congenita: Keratosis disseminata 130(2), 425Y429. doi: 10.1038/jid.2009.215.

VOLUME 5 | NUMBER 1 | JANUARY/FEBRUARY 2013 47

Copyright © 2013 Dermatology Nurses' Association. Unauthorized reproduction of this article is prohibited. achyonychia congenita (PC) is a . rare skin diso rder caused by muta Ptions affectmg a group of keratins found in specific regions of the epi dermis. Although originally classified as a nail dystrophy, its major impact on adult patients is from painful plan tar keratoderma. Other manifestations also commonly occur. This article re views the clinical presentation, genetic diagnosis, pathogenesis and current and future treatment options for Pc.

CLINICAL PRESENTATION Data ga thered by the International PC Research R egistry (IPCRR) on n1.ore than 600 individuals with geneti ca lly confirmed PC show that PC al ways includes at least lather feature

along with nail dystrophy. I The most debilitating condition for those with PC is exquisitely painful plantar keratoderma. In some cases, this is the predominant clinical presentation ' with little or no nail dystrophy. 2.3 Other CO lTIJnon manifestations of PC are ex tensive cys ts including steatocysts, leu kokeratosis, follicular hyperkeratosis and palmar keratoderma (Figures 1-7).1 However, only thickened nails may be evident in neonates and infants with PC4 Natal teeth may also be present at birth, predominantly in one PC subtype. Leukokeratosis is another finding for PC infants and is often misdiagnosed and treated as thrush. Although florid leukokeratosis has often been incrimi nated as the cause of difficulty in feeding during il"lfancy, especially with laryngeal • involve ment, an alternative underlying ca use is a painful "first bite sy ndrome" experienced by some PC children. This Pachyonych la is being assessed in a se ries of unpub lished cases in the IPCRR. In each of these patients, a change to soft rupples with large holes inU11.ediately resolved Congenita the feeding problem. Blisters or callus on the feet usually begin when a child with PC fIrSt begins to walk and the age of onset of plantar Overview keratoderma is a fimction of the extent of weight on the feet. The pain is usually The inherited nail dystrophy is a ~s ociated with painful plantar keratoderma. constant by age 10.4 Many adults with PC rely on canes, crutches or wheelchairs or MARY E. SCHWARTZ, LLD , C DAVID HANSEN, MD, AMY S. PALLE R, MD, even resort to crawling on their knees to FRANCESJ D. SMITH , PHD, AND ELI SPRECHER, MD, PHD avoid increased plantar pain by walking. Utilizing ultrasound linages, research ers have recently captured images of the

36 April 2014 I THE DE RMATOLOGIIT" I www.the-dermotologist.com PACHYONYCHIA CONGENITA OVERVIEW

achyonychia congenita (PC) is a underlying plantar blisters in PC patients. rare skin disorder ca used by muta T hese are not found in other asymptom Ptions affecting a group of keratins atic calluses of patients witll other condi found in specifi c regions of the epi tions that have a similar appearance. This dermis. Although originally classified study has contributed to our understand as a nail dys trophy, its major impact ing of the source of the incessant pain as on adult pati ents is from painful plan sociated with PC plantar keratoderma s tar keratoderma. Other manifes tations From age 4 to 14 years, children with also commonly occur. This article re PC may experience extensive follicu views the clinical presentation, genetic lar hyperkeratosis especially in areas of diagnosis, pathogenesis and current friction around the waist, knees and el and future treatment options for Pc. bows. In some types of PC, cysts are the dominant feature, characterized by CLINICAL PRESENTATION milia in infa ncy and childhood and ex D ata gathered by the International tensive body cysts at puberty continu PC Research Registry (IPCRR) on ing throughout adult life. more than 600 individuals with geneti cally confirmed PC show that PC al PATHOGENESIS ways includes at least 1 other feature Keratins are structural proteins that along with nail dys trophy. I promote the integrity of epithelial cells. The most debilitating condition for As a result mutations in the genes en those with PC is exquisitely painful coding keratins lead to cell fragility6.7 plantar keratoderma. In some cases, this The skin expresses the largest number of is the predorninant clinical presentation ' keratin genes of any organ. Widely dis with little or no nail dystrophy.2.3 Other tributed lesions in keratin disorders, as conunon manifestations of PC are ex occurs in epidermolysis bullosa simplex, tensive cysts including steatocysts, leu result £i.-om mutations in genes expressed kokeratosis, follicular hyperkeratosis and throughout the epidermis. P C is ca used palmar keratoderma (Figures 1-7).1 by mutations in 5 keratin genes KRT6a, However, only thickened nails may be KRT6b, KRT6c, KRT16 and KRT1 7 evident in neonates and infants with w hich are expressed only in palmoplan pc.4 N atal teeth may also be present at tar skin, the nail bed, pilosebaceous unit birth, predominantly in one PC subtype. and oral mucosa, leading to selective in Leukokeratosis is another finding for volvement of these sites in pc. 3.7.8 PC infants and is often misdiagnosed and treated as thrush. Although florid GENETIC DIAGNOSIS leukokeratosis has often been incrimi PC is an autosomal-dominant diso r nated as the ca use of difficulty in feeding der, which has been reported worlwide during infancy, especially with laryngeal with approximately equal prevalence involvement, an alternative underlying in males and females. M ore than 45% cause is a painful "first bite syndrome" of cases appear spontaneously with no experienced by some PC children. This fa mily history of pc. I Given the over Pachyonychia is being assessed in a series of unpub lapping clinical presentation with other lished cases in the IPCRR. In each of geneti c disorders, only geneti c testing these patients, a change to soft nipples can confirm the P C diagnosis. Congenita with large holes immediately resolved With nearly 100 distinct PC muta the feeding problem. tions now identified, correlating the Blisters or callus on the feet usually signs of PC with specific mutations begin when a child with PC first begins and geiles has led to a new classifica Overview to walk and the age of onset of plantar tion sys tem of p c. While in the past, keratoderma is a function of the extent PC has been classified according to of weight on the feet. T he pain is usually phenotypic features into PC-l and PC- The inherited nail dystrophy is associated with painful plantar keratoderma. constant by age 10. 4 Many adults with PC 2, the disorder is now classified into 5 rely on canes, crutches or wheelchairs or subgroups corresponding to the under IWRY E. SCHWARTZ, LLD, C. DAVID HANSEN, MD, AMY S. PALLER , MD, even resort to crawling on their knees to lying genetic defect: PC-K6a, PC-K6b, 8 FRANCES J. D. SMITH, PHD, AND ELI SPRECHER, MD, PHD avoid increased plantar pain by walking. PC-K6c,PC-K16 and PC-K17.'· Utilizing ultrasound images, research As treatment development is fo cused Figures 1-7. Typical characteristics of pachyonychia congenita include (0) facial cysts (b and c) plantar keratoderma (d) ers have recently captured images of the on specific genes and mutations, free ge- leukokeratosis (e) follicular hyperkeratosis (I) fingernail dystrophy (g) toenail dystrophy.

36 April 2014 I THE DERMATOLOGllTo I www.the-dermatologist. com Apri.1 2014 I THE DERMATOLOGllTo I www.the-dermatologist.com . 37 PACHYONYCHIA CONGENITA OVERVIEW netic testing is provided to each patient In a clinical trial, inj ection into plan Dr. Paller is a pediatric dermatologist at through the IPCRR sponsored by the tar skin of a small intelfering RNA Northwestern University in Ch icago} IL. PC Project (www.pachyonychia.org). (siRNA) that specifically suppressed Dr. Smith is a senior research fe llow at the Genetic testing not only confirms di mutant KRT6a reduced the kerato University ifDund ee in Dllndee} Scotland. agnosis but also aids geneti c counseling. derma at the inj ection site.12 H owever, Dr. Sprecher is director if the dermatology the pain of treatment highlighted the departlnent at Te IAviv SoU/'asky Medical Cen TREATMENT need for alternative delivery methods tre in Tel Aviv, Israel. Currendy there is no specific therapy particularly because such trea tment re for PC so th e mai n aim of treatment is quires regular application to sustain the Disclosure: The authors report no relevant finan to alleviate the pain ca used by the plan clinical res ult. cial relationships. tar keratoderma. M echanical methods to A clinica l trial of siRNA, KRT6a References 9 remove the callus are most effective. The delivered by micro needles, is under 1. Eliason MJ, Leachman SA, Feng BJ, Schwartz " not too thick, not too thin" motto is es going regulatory approval. R apa my ME, Hansen Co. A review of th e cl inical pheno sential in caring for keratoderma. Bleach cin has been shown to inhibit the type of 254 patients with geneti ca ll y confirmed baths ca n reduce the onset of infections. translation of K6a mRNA but has pac hyo nyc hia congenita. ) A lii Acad Derlllatol. 2012;67(4):680-686. Many conUl10n treatments recom significant toxicity in its oral formu 2. Smith FJ , Fisher MP, Healy E, et a!. Novel 13 mended by dermatologists (such as urea Iation. A trial o f topically-applied keratin 16 mutations and protein expression or salicylic acid- based emollients) are rapamycin for the plantar keratoder studies in pachyo nyc hia conge ni ta type 1 and not useful in managing the PC kerato- ma is under way. focal paLnoplantar keratoderma. Exp Derlllato/. 2000;9(3): 170-177. 3. Wilson NJ, Messenger AG, Leachman SA, et a!. Keratin K6c mutations ca use foca l pal Currently there is no specific therapy for PC so the main aim moplantar keratoderma. ) il lIJest Derlllato /. of treatment is to alleviate the pain caused by th~ plantar 2010;130(2):425-4299. 4. Shah S, Boen M, Kenner-Bell B, Schwartz M, keratoderma. Mechanical methods to remove the callus are Rademaker A, Paller AS. Pachyonyc hia congenita most effective. The "not too thick, not too thin" motto is in pediatric patients: natural history, features, and essential in caring for keratoderma. Bleach baths can reduce impact.jAJ\.1A Derlllato/. 2014;150(2): 145-153. 5. Goldberg I, Sp recher E, Schwa rtz ME, Gaitini the onset of infections. 0. Comparati ve study of high- resolution multifre quency ultrasound of the plantar skin in patients with va rious types of hereditary palm oplantar keratoderma. Derlll atology. 2013;226(4):365-370. derma and al:e inferi or to at least weekly CONClUSION 6. Coulombe PA , Lee C H . Defining keratin pro trinmling in most types of Pc. Retinoid PC is a rare genetic diso rder for tein fun ction in skin epitheli a: epidermolysis bul treatments, effective fo r some of the w hich there are ve ry few therapeutic losa simplex and its aftermath.) fl lIJes t Derlllatol. other keratin disorders, are also inef options. By building a patient com 2012;132(3 Pt 2):763-775. fective for PC plantar keratoderma and munity through the IPCR..R and a 7. McLean WH, Hansen CD, Eli ason MJ, Smith FJ. The phenotypic and molecular genetic fea may increase pain.lo Mechanical trea t physician and researcher community tures of pachyonychia congeni ta.) It west Derlllatol. ment for nails is also favored and surgi through the IPCC, PC Project is mov 20 11 ;131(5): 1015-10 17. cal removal of cysts is often required.9 ing research forward to better under 8. Wilson NJ, Leachman SA, Hansen CD, et a!. A An important aspect of pati ent ca re stand the condition and to develop ef large mu tational study in pac hyo nychi a congenita. is recognizing the feelin g of isolation fe ctive trea tments. ) It llJes t Denl/otol. 2011;13J (5): 1018-1024. 9. Goldberg I, Fruchter 0 , Meili ck A, Schwartz ass ociated with this rare and highly vis Working alone, a single patient or ME, Sprecher E. Best treatment practices for ible skin disorder. Encouraging patients physician ca nnot solve the ques tions pac hyo nyc hia congenita.j Ellr Acad Denl/otol Vel/e to becom e part of the IPCRR can of a rare disease. Sinlilarly experimen reol.2014;28(3) :279-285. provide tremendous emotional support tal trials on individual pati ents will 10. Gruber R , Edlinger M , Kaspar R.L, et al. An because it connects them with others not yield the data l:equired to set stan appraisal of oral retinoids in the trea tment of pachyo nyc hia congenita . ) A lii Acad Derlllotol. w ho understa nd the pain of PC and dards of practi ce to provide effective 2012;66(6):eI 93-eI99. the functional and psyc hological affect treatment to aU P C patients. Only by 11 . Sc hwartz ME, Z inunerman GM, Smith FJ, of the diso rder. II This can be especially working together can we make ~ dif Sprecher E. Pac hyo nchia Conge nita Project: a important for patients who have no ference.We invite all patients, research partnership of patient and medical professional.) other affected fa mily members. ers and dermatologists to j oin with us Derlllatol N ll rses} Assoc. 2013;5(,1) :42-47. 12. Leachman SA, Hickerso n RP, Schwa rtz ME, More effective treatments are under in a cooperative effort to make prog et a!. First-in- human mutation-targeted siRNA inves tigation in preclinical and clinical ress in finding effective trea tments for phase Ib trial of an inherited ski n disorder. Mol trials. Physician members of the Inter this painful condition . • Ther. 2010;18(2):442-446. national PC Conso rtium (IPCC) have 13. Hi ckerson RP, Leake D, Pho LN, Leachman access to the lates t PC research results Mary Schwartz is director if Pachyonychia SA, Kas par R.L. Rapamycin selectively inhibits expression of an inducible keratin (K6a) in hu and will be invaluable partners for PC Congenita Project in Salt Lake CitJ\ UT man ke ratinocytes and improves symptoms in patients as therapies are emerging and D,: Han sen is a dermatologist at the Univer pachyo nyc hia conge nita patients. ) Derlll otol Sci. clinical trials results become ava ilable. sity if Utah in Salt Lake Cit)\ UT 2009;56(2) :82-88.

38 April 2014 I THE DERMATOLOGIIT' I www.the-dermatol?gisl.com

Research

Original Investigation Pachyonychia Congenita in Pediatric Patients Natural History, Features, and Impact

Sonal Shah, MD; Monica Boen, MD; Brandi Kenner-Bell, MD; Mary Schwartz, PhD; Alfred Rademaker, PhD; Amy S. Paller, MD

Supplemental content at IMPORTANCE Nail dystrophy in early childhood often suggests a diagnosis of pachyonychia jamadermatology.com congenita (PC). No previous investigation has focused on the early signs of PC and the natural course of the disease.

OBJECTIVES To determine the course of pediatric PC, correlate the disease course with the clinical appearance and specific gene mutations, and assess the effect of pediatric PC on quality of life.

DESIGN, SETTING, AND PARTICIPANTS One hundred one patients or families with genetically confirmed PC from the International Pachyonychia Congenita Research Registry who completed a survey on the general clinical features of PC and an auxiliary questionnaire on the clinical presentation and quality-of-life issues related to pediatric PC.

EXPOSURE Individuals with pachyonychia congenita.

MAIN OUTCOMES AND MEASURES Completion of both surveys.

RESULTS At birth, toenail changes were present in 47.5% of patients; fingernail changes in 40.6%; and plantar keratoderma in 6.9%. By 5 years of age, these 3 key manifestations were found in 81.2%, 74.2%, and 75.3%, respectively, of individuals with genotype-confirmed PC. The correct diagnosis was made during the first year of life in 26.7% of patients despite the presence of toenail dystrophy in more than 65.3%. Clinical differences that distinguished PC subtypes included (1) later onset and less frequent occurrence of nail dystrophy and keratoderma in PC-K6b, PC-K6c, and PC-K16; (2) concurrent fingernail and toenail thickening in PC-K6a and PC-K17; (3) more palmar keratoderma in PC-K16; (4) cysts primarily in PC-K17 and follicular hyperkeratoses primarily in PC-K6a; (5) hoarseness and/or oral leukokeratoses in the first year of life most often in PC-K6a; and (6) natal teeth exclusively in PC-K17. Among pediatric patients, PC affected the social interactions and function of adolescents most profoundly.

CONCLUSIONS AND RELEVANCE Among patients with a detectable mutation, PC manifests Author Affiliations: Department of with nail thickening and plantar keratoderma before school age in more than three-quarters Dermatology, Northwestern of affected children, allowing early diagnosis. The highly visible nail changes and painful University Feinberg School of plantar thickening exert a psychosocial effect on most affected adolescents. Medicine, Chicago, Illinois (Shah, Phenotype-genotype correlations in children with PC validate the new classification based on Boen, Kenner-Bell, Paller); Department of Pediatrics, the affected gene. Northwestern University Feinberg School of Medicine, Chicago, Illinois (Kenner-Bell, Paller); Pachyonychia Congenita Project, Salt Lake City, Utah (Schwartz); Department of Preventive Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois (Rademaker). Corresponding Author: Amy S. Paller, MD, Department of Dermatology, Northwestern University Feinberg School of Medicine, 676 N St Clair, Ste 1600, JAMA Dermatol. doi:10.1001/jamadermatol.2013.6448 Chicago, IL 60611 (apaller Published online October 16, 2013. @northwestern.edu).

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Research Original Investigation Pachyonychia Congenita in Pediatric Patients

achyonychia congenita (PC) constitutes a group of al- An auxiliary 48-question survey (approved by the IRB at Ann most exclusively autosomal dominant disorders of and Robert H. Lurie Children’s Hospital of Chicago) was sent by P paired keratins of the nails and skin. Since PC was ini- the Pachyonychia Congenita Project to 254 registry families with tially reported in 1906 by Jadassohn and Lewandowsky,1 more genotype-confirmed PC (Supplement [Appendix]) and ad- than 500 cases have been registered or otherwise described. dressed pediatric-specific issues, early signs, and the natural his- Pachyonychia congenita manifests as nail dystrophy, painful tory of PC. Parents completed the survey for preteenaged pa- focal palmoplantar keratoderma, follicular keratoses, muco- tients with input from children as appropriate; patients of all ages sal leukokeratoses, hoarse voice, cystic lesions, and, rarely, na- were encouraged to consult with their parents to answer ques- tal teeth.2,3 tions related to the disorder early in life. Informed consent was Underlying keratin gene mutations have been described in obtained from all participants. The results of both question- the 5 keratin genes, KRT6A (OMIM 148041), KRT6B (OMIM naires were sent to the Pachyonychia Congenita Project, which 148042), KRT6C (OMIM 612315), KRT16 (OMIM 148067), and sorted data by keratin mutation and provided de-identified data KRT17 (OMIM 148069), which alter keratins 6a, 6b, 6c, 16, and for compilation and statistical analysis at Northwestern Uni- 17, respectively.4-9 These genes are expressed in the nail bed, versity. We performed Fisher exact testing to determine statis- palmoplantar epidermis, and mucosa. Keratins play a key role tical significance (defined as P < .05) in comparing clinical fea- in epidermal cell integrity and mechanical strength. Muta- tures with underlying keratin gene mutations. All statistical tions in these 5 keratin genes cause epidermolysis and com- analyses were performed with commercially available soft- pensatory hyperkeratosis at these sites. Historically,2 major sub- ware (SPSS, version 15.0 [SPSS Inc], and Stata, version 10.0 types were based on clinical characteristics. The Jadassohn- [StataCorp]). Because PC-K6c was first termed PC after the aux- Lewandowsky PC type 1 (PC-1) often showed associated oral iliary questionnaire was introduced (2012), only data from the leukokeratoses,1 and the Jackson-Lawler PC type 2 (PC-2) of- first questionnaire were available from patients with this sub- ten showed cysts and occasionally natal teeth.3,10 Pachyo- type. Given its mild clinical features, PC-K6c was excluded in nychia congenita type 1 was originally linked to mutations in the statistical analyses to distinguish among PC subtypes. Data from type II keratin gene KRT6A and its type I expression partner individuals with mutations in the GJB6 gene (OMIM 604418), KRT16 and PC-2 with mutations in KRT6B and KRT17.3,4,6,8,11 also known as connexin 30, were derived from the original Pachyonychia congenita has been genotyped at no cost in questionnaire. individuals who enroll in an international registry, enabling comprehensive genotype-phenotype analysis.12-16 Pheno- typic overlap among PC genotypes has now made obsolete the Results designations of PC-1 and PC-2. Instead, a new classification based on mutations in the keratin-encoding genes KRT6A, Of 254 patients enrolled in the IPCRR who completed the origi- KRT6B, KRT6C, KRT16, and KRT17 divides PC into subtypes nal questionnaire and had genetically confirmed PC, 101 re- PC-K6a, PC-K6b, PC-K6c, PC-K16, and PC-K17, respectively.16 sponded to the addendum questionnaire (response rate, 39.8%). Features of PC usually manifest during the first 3 years of The demographic features of these 101 PC patients and the 8 PC- life,12 allowing the diagnosis to be considered. However, little K6c patients who responded to the original questionnaire are attention has been paid to the disease course, early diagnos- described in Table 1. Of the returned questionnaires, 78.0% were tic features, and effect on quality of life of PC in children. To completed by adult patients who were able to report the onset facilitate early diagnosis and increase our understanding about of features and impact of PC throughout the first 18 years of life, the impact of PC in children, affected families and patients were whereas 22.0% of the returned questionnaires reflected the ex- polled about pediatric-specific issues. perience to date of affected individuals currently in their first 2 decades of life. Among the respondents, 42.2% had mutations in KRT6A (PC-K6a), and an additional 28.4% had mutations in Methods KRT16 (PC-K16). Almost 60% in the registry had a family history of PC, reflecting the high rate of spontaneous mutation in Two questionnaires were administered to families participat- keratin genes. The diagnosis of PC by a physician was made being in the International Pachyonychia Congenita Research Reg- fore 1 year of age in 27 patients (26.7%, of whom 12 [44.4%] had istry (IPCRR) through the Pachyonychia Congenita Project (ww- a known family history) despite the presence of toenail and fin- w.pachyonychia.org). The study and questionnaires were gernail dystrophy in approximately 60% of the affected in- approved by Western Institutional Review Board (WIRB Study fants (see below). By 5 years of age, the correct diagnosis was number 1057496; protocol number 20040468). The first ques- made in 43 patients (42.6%, of whom 18 [41.9%] had a known tionnaire asked general questions about PC features, their ef- family history) despite toenail dystrophy in 81% (82 of 101), fin- fect on daily life, and interventions after registry enrollment12 gernail dystrophy in 74% (75 of 101), and plantar keratoderma (Supplement [eAppendix]). The diagnosis of PC was con- in 75% (76 of 101). firmed by dermatologists from the Pachyonychia Congenita Project medical and scientific advisory board through assess- Toenail Dystrophy ing family-provided questionnaire data, photographs of skin The earliest and most common clinical characteristic of PC was and nail changes, and a 1-hour patient/family telephone con- toenail dystrophy (Figure 1A), noted to involve at least 2 dig- sultation. its in 98.2% of patients by the time of reporting; the only ex-

E2 JAMA Dermatology Published online October 16, 2013 jamadermatology.com

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Pachyonychia Congenita in Pediatric Patients Original Investigation Research

Table 1. Demographics of Registry Patients and Distinguishing Clinical Features Among PC Subtypesa

PC Subtype All PC-K6a PC-K6b PC-K6c PC-K16 PC-K17 (N = 109 (n=46 (n=10 (n=8 (n=31 (n=14 Feature [100.0%])b [42.2%]) [9.2%]) [7.3%]) [28.4%]) [12.8%]) P Valuec Demographic Current age, y Mean (SD) 36.0 (19.9) 32.4 36.1 37.6 43.6 30.1 >.05 Median (range) 35.5 (0-81) 33 41 44 43 27 Age group, y (0-63 y) (9-54 y) (6-72 y) (1-81 y) (3-68 y) 0-5 9 (8.3) 5 (10.9) 0 0 1 (3.2) 3 (21.4) 6-10 8 (7.3) 4 (8.7) 1 (10.0) 2 (25.0) 0 1 (7.1) 11-17 7 (6.4) 2 (4.3) 1 (10.0) 0 3 (9.7) 1 (7.1) ≥18 85 (78.0) 35 (76.1) 8 (80.0) 6 (75.0) 27 (87.1) 9 (64.3) Sex Female 61 (56.0) 26 (56.5) 2 (20.0) 5 (62.5) 17 (54.8) 11 (78.6) >.05 Male 48 (44.0) 20 (43.5) 8 (80.0) 3 (37.5) 14 (45.2) 3 (21.4) Family history positive for PC 65 (59.6) 21 (45.7) 8 (80.0) 8 (100.0) 20 (64.5) 8 (57.1) >.05 Keratin gene mutation KRT6A KRT6B KRT6C KRT16 KRT17 >.05 Clinical Dystrophy Toenail 107 (98.2) 46 (100.0) 10 (100.0) 6 (75.0) 31 (100.0) 14 (100.0) >.05 Fingernail 84 (77.1) 46 (100.0) 4 (40.0) 0 21 (67.7) 13 (92.9) <.001 Keratoderma Plantar 107 (98.2) 45 (97.8) 10 (100.0) 8 (100.0) 31 (100.0) 13 (92.9) >.05 Palmar 47 (43.1) 17 (37.0) 4 (40.0) 0 21 (67.7) 5 (35.7) <.05 Paind Plantar 100 (91.7) 43 (93.5) 10 (100.0) 8 (100.0) 30 (96.8) 9 (64.3) <.05 Palmar 23 (21.1) 11 (23.9) 1 (10.0) 0 9 (29.0) 2 (14.3) <.05 Oral leukokeratosis 81 (74.3) 45 (97.8) 4 (40.0) 0 17 (54.8) 5 (35.7) <.001 Cyst 71 (65.1) 41 (89.1) 7 (70.0) 1 (12.5) 9 (29.0) 13 (92.9) <.001 Follicular hyperkeratosis 54 (49.5) 37 (80.4) 6 (60.0) 0 4 (12.9) 7 (50.0) <.001 Natal teeth 14 (12.8) 2 (4.3) 0 0 0 12 (85.7) <.001 Hoarseness 17 (15.6) 14 (30.4) 0 0 1 (3.2) 2 (14.3) <.05

Abbreviation: PC, pachyonychia congenita. c Statistical analysis compares subtypes PC-K6a, PC-K6b, PC-K16, and PK-17. a Unless otherwise indicated, data are expressed as number (percentage) of P < .05 indicates statistical significance. patients. Percentages have been rounded and may not total 100. d Indicates percentages with pain among patients with plantar or palmar b Includes the 101 patients who responded to both questionnaires and the 8 keratoderma. PC-K6c patients who responded to the original questionnaire only.

ception was absence of toenail dystrophy in 2 of the 8 PC-K6c same time; patients with PC-K6a were more likely to have all patients (25.0%) (Table 1). Of the patients with toenail involve- 10 toenails affected at onset than were patients with PC-K16 ment, all 10 toenails were affected in 74.3%, 98.0% had in- or PC-K6b (P < .001). The hallucal nail was the most common volvement of toenails of the fifth digit, and 98.0% had changes of the toenails to first become affected (89.1% of PC patients of the hallucal nails by the time of reporting. Toenail abnor- with toenail dystrophy). The most common initial toenail malities were present at birth in 39.0% of respondents over- change was thickening (77.2% of all patients), and 65.3% also all, although they appeared most often in neonates with PC- showed toenail discoloration at onset. Toenail thickening pro- K6a and PC-K17 (P < .001) (Supplement [eFigure, A]). By 1 and gressed throughout the first decade of life in most of the pa- 5 years of age, nail dystrophy was noted in 65.3% and 80.2% tients. The nail dystrophy occurred before 6 years of age in 5 of all respondents, respectively. By 5 years of age, all 46 chil- of the 6 PC-K6c patients with this feature (83.3%), but the dys- dren with PC-K6a, the most common subset, showed toenail trophy was mild, affecting only 1 toenail bilaterally (usually the changes. In contrast, toenail changes did not appear in chil- fifth toenail) or, in 1 case, multiple toenails unilaterally. The 2 dren with PC-K6b until at least 1 year of age and progressively PC-K6c patients without nail involvement were adults. increased in occurrence thereafter. In 99.0% of patients with Fingernail changes occurred overall in 76.1% of patients toenail dystrophy, more than 1 nail became dystrophic con- and, as with toenail changes, were most severe in PC-K6a. Over- currently, and 70.3% had all 10 nails become dystrophic at the all, 40.6% of patients had fingernail changes at birth. Finger-

jamadermatology.com JAMA Dermatology Published online October 16, 2013 E3

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Research Original Investigation Pachyonychia Congenita in Pediatric Patients

Figure 1. Representative Images of Clinical Features of Pachyonychia Congenita and Genotype in Affected Children

A B

C D

A. Subungual hyperkeratosis and V-shaped thickening of the toenails in a toddler with a KRT17 M88K mutation. B. Discolored and thickened distal aspect of all fingernails owing to a KRT16 R127P mutation. C, Severe plantar keratoderma in a child with a KRT16 S130del mutation. D, Mild plantar keratoderma in a child with a KRT6C E472K mutation. E, Oral E F G leukokeratosis of the tongue in an infant with a KRT6A L468P mutation. Leukokeratosis was misdiagnosed initially as a candidal infection. F, Oral leukokeratoses of the tongue in an adolescent with a KRT6A N172del mutation. G, Follicular keratoses on the knee owing to a KRT6A N172del mutation.

nail involvement occurred in most of the neonates with PC- patients with fingernail dystrophy, regardless of PC subtype, K6a and PC-K17 but infrequently in neonates with PC-K16 and had toenail involvement. However, patients with PC-K6a and never in neonates with PC-K6b (P < .001) (Supplement [eFig- PC-K17 were more likely to develop dystrophy of their finger- ure, B]). Fingernail changes never developed in 5 patients with nails and toenails concurrently, whereas patients with PC- PC-K6b (50.0%), 11 patients with PC-K16 (35.5%), 1 patient with K6b and PC-K16 were more likely than other subtypes to de- PC-K17 (7.1%), and the 8 patients with PC-K6c (100.0%). By 5 velop toenail dystrophy first (P < .001). years of age, all children with PC-K6a and 13 children with PC- The toenails and fingernail dystrophy varied in appear- K17 (92.9%) showed fingernail dystrophy, in addition to toe- ance from thickened, shortened, friable nails to nails with nail dystrophy. Fingernail dystrophy developed simultane- marked subungual thickening and a typical pinched, V- ously in all 10 nails in 68.7% of PC patients; only 6.0% reported shaped curvature (Figure 1A and B). We found no correlation involvement in only 1 nail, with no consistency as to the in- between the appearance of the nails and the PC subtype. volved digit. The initial change in fingernails was nail thick- Periungual infections occurred in 76.6% of PC patients ening in 65 of the 86 patients with nail changes (75.6%), usu- overall but more often in those with PC-K6a (43 of 46 [93.5%]) ally in combination with changes in color (41 [63.1%]). All than those with other subtypes (P < .05). Most patients with

E4 JAMA Dermatology Published online October 16, 2013 jamadermatology.com

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Pachyonychia Congenita in Pediatric Patients Original Investigation Research

nail infections (52 of 73 [71.2%]) never sought treatment by a developed hoarseness, 16 (93.3%) concurrently showed oral physician. Most of these patients treated nails by soaking them leukokeratoses. Natal teeth virtually clinched the diagnosis (38 of 61 [62.3%]), lancing purulent areas or trimming the nails of PC-K17 (86.0% [P < .001]) but were noted in 2 of 46 infants (35 of 61 [57.4%]), or using topical medication (28 of 61 [45.9%]). (4.3%) of infants with PC-K6a. The teeth were described as Nail infection or its clearance did not lead to a change in ap- soft or crumbly and were rapidly lost or were described as pearance of the dystrophic nail. normal in appearance and persistent until the deciduous teeth erupted. Painful Plantar Keratoderma The feature of PC that has the most profound effect on qual- Cysts ity of life is painful plantar keratoderma (Figure 1C and D).12 Cysts were noted overall in 69.3% of patients with PC. Most Although present at birth in fewer than 10% of individuals with patients who developed cysts had PC-K6a, PC-K6b, or PC- PC, 24.8% of PC patients overall noted plantar keratoderma by K17, without a statistically significant difference in the risk for 1 year of age, 75.3% by 5 years of age, and 89.1% within the first developing cysts among these subgroups. Patients with the PC- decade of life (Supplement [eFigure, C]). Of the 27 patients di- K16 subtype were the least likely to develop cysts (P < .05), and agnosed with PC before 1 year of age, 12 (44.4%) had plantar cysts did not develop in PC-K6c patients. The onset of cysts keratoderma by the time of diagnosis. In patients with PC- occurred most often from 6 to 10 (31.6%) and 11 to 20 (36.8%) K6a, PC-K16, and PC-K17, the onset of plantar keratoderma usu- years of age. Patients primarily treated their cysts by lancing ally occurred before age 5 years, whereas in patients with PC- them (37 of 54 [68.5%]) or applying topical antibiotic oint- K6b and PC-K6c, onset was usually after age 5 years. The most ment (26 of 54 [48.1%]). common initial locations of the plantar keratoderma were at pressure points on the heel (66 of 98 [67.3%]) and ball (63 of Follicular Hyperkeratoses and Hyperhidrosis 98 [64.3%]) of the foot. During the first decade of life, 70 of Follicular hyperkeratoses were described in more than half of the 73 patients with keratoderma (95.9%) had plantar pain, the PC patients, although only in 26 (25.7%) by preschool age which compromised their function; pain occurred later in chil- (Figure 1G). They occurred most often in PC-K6a (80.4%) and dren with PC-K6b than the other subtypes (P < .05). Most pa- PC-K6b (42.9%) and least commonly in PC-K16 (12.9%; P < .001 tients with plantar keratoderma also had local skin infections compared with PC-K6a). In patients with PC-K6a, 14 of 37 (47 of 99 [47.5%]). (37.8%) developed cysts at 1 to 5 years of age and 23 of 37 (62.1%) Palmar keratoderma occurred in only 45.5% of patients showed follicular hyperkeratoses by the second decade of life overall by the time of the response to the questionnaire. Of (Supplement [eFigure, E]). Only 1 PC-K16 patient developed those who developed palmar keratoderma, 24 of 47 (51.1%) saw follicular hyperkeratoses during childhood, and 27 of 31 pa- changes by 5 years of age, and 32 of 47 (68.1%) by 10 years tients with PC-K16 (87.1%) never developed follicular hyper- (Supplement [eFigure, D]). Palmar keratoderma was more of- keratoses by the time of reporting. Hyperhidrosis was de- ten reported in PC-K16 than in any other subtype (67.7%; scribed in 51.5% of respondents of all ages, but in only 5 of 22 P < .05) and was often complicated by painful erosions, bul- children (22.7%). Alopecia was not an issue in children with lae, or fissures of the palms, usually before 5 years of age. PC in the first decade of life, and its association with PC is ques- tionable. In the original survey of 254 PC patients, 15 patients Other Clinical Manifestations ranging in age from 15 to 81 years had hair loss. Of these pa- Although PC-K6b and PC-K6c were often distinguishable from tients, 10 were male and most had no other hair abnormality, other forms because of their more limited manifestations, fea- raising the question of androgenic alopecia as the cause. All tures other than nail dystrophy and keratoderma were help- but 1 of these individuals had PC-K6a or PC-K16. ful in differentiating among the PC-K6a, PC-16, and PC-17 subtypes in children. In particular, oral leukoplasia, cysts, and Effect on Quality of Life in Adolescents keratoses varied by PC subtype. Plantar keratoderma in PC was characteristically painful by the second decade of life and led to the greatest effect on quality Oral Leukokeratoses of life. More than 50% of children found that the plantar kera- The oral leukokeratoses occurred in 70.3% of PC patients and toderma affected walking and playtime; however, less than half were more strongly associated with PC-K6a than with any other of patients also found impedance with crawling, schoolwork, subtype (P < .001). Of those affected, the median age at onset and chores (Figure 2A). The effect on function more often be- was 3 weeks, and 26 of 71 (36.6%) experienced a first occur- gan after 5 years of age, peaked during adolescence, and oc- rence during the first year of life. Oral leukokeratoses were of- curred most often with PC-K6a. Pachyonychia congenita af- ten mistaken for thrush (Figure 1E) but failed to respond to an- fected the social life of virtually all patients of school age or ticandidal therapy. Oral leukokeratoses were most commonly older. Most experienced limitations related to wearing clothes noted on the tongue (Figure 1E and F) (68 of 71 [95.8%]). and playing sports (Figure 2A), experienced teasing, and were Hoarseness was noted in 12.9% of patients overall, and PC- embarrassed by their nails, especially during adolescence K6a patients were more likely to develop hoarseness than PC- (Figure 2B and Table 2). Of the 84 PC patients with fingernail K6b and PC-K16 patients (P < .05) but not PC-K17 patients. The involvement, 66 (78.6%) concealed their nails, especially by onset of hoarseness was variable but always present before 3 keeping the fingers curled (50 [59.5%]) or crossing the arms (46 years of age in the affected patients. Of the PC patients who [54.8%]). Other techniques included keeping hands in the pock-

jamadermatology.com JAMA Dermatology Published online October 16, 2013 E5

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Research Original Investigation Pachyonychia Congenita in Pediatric Patients

Figure 2. Functional Impairment and Extent of Embarrassment by Pachyonychia Congenita (PC) Subtype

A 2.5 PC-K6a PC-K6b 2.0 a PC-K16 PC-K17 1.5

1.0

Frequency of Impairment Frequency 0.5

0.0 Crawling Walking Sports Playtime School Chores B Activities of Daily Life 2.5 PC-K6a a PC-K6b 2.0 PC-K16 PC-K17 1.5

1.0 A, Effects of plantar or palmar keratoderma on activities of daily life. B, Age at embarrassment about the 0.5 appearance of nails. Given its mild

Mean Frequency of Embarrassment Mean Frequency clinical features, the PC-K6c subtype was excluded from the analysis. 0.0 aPatients or families graded 1-5 6-10 11-20 21-30 >30 frequency on a 4-point scale, where Age Group, y 0 indicates never; 1, little effect; 2, moderate effect; and 3, always/worst burden.

Table 2. Effects of PC on Social Life of School-aged Children

PC Subtype, No. (%) Reporting an Adverse Effect on Social Lifea All, No. (%) PC-K6a PC-K6b PC-K16 PC-K17 Effect (n = 92) (n = 41) (n = 10) (n = 30) (n = 14) P Valueb Teasing 69 (75.0) 37 (90.2) 4 (40.0) 20 (66.7) 8 (72.7) <.01 Difficulty wearing certain clothes/accessories 79 (85.9) 36 (87.8) 5 (50.0) 29 (96.7) 9 (81.8) <.01 Difficulty participating in recreational 63 (68.5) 36 (87.8) 6 (60.0) 26 (86.7) 8 (72.7) .13 activities Dating problems 40 (43.5) 22 (52.6) 3 (33.3) 12 (40.0) 5 (45.5) .68

Abbreviation: PC, pachyonychia congenita. b Statistical analysis compares subtypes. P < .05 indicates statistical a None of the patients with PC-K6c reported an adverse effect on their social significance. life.

ets (31 [36.9%]), using nail polish (25 [29.8%]), using artificial connexin 30 mutations, but not PC, were hearing loss and alo- nails (9 [10.7%]) and wearing gloves (7 [8.3%]). pecic patches or sparseness, with hair that was described as thin and sometimes brittle. PC-Like Early Nail Dystrophy and Connexin 30 Mutations Only patients with gene-confirmed PC subsets completed the pediatric-specific questionnaire. However, among patients in Discussion the original registry questionnaire, 7 with nail dystrophy and clinically presumed PC had mutations in the connexin 30 gene Pachyonychia congenita constitutes a group of primarily auto- (GJB6)(Table 3). Of these, 5 (71.4%) had abnormalities of the somal dominant inherited disorders caused by mutations in toenails and fingernails at birth, with the other 2 developing paired keratins. Genotyping has been provided since 2004 as a dystrophy of the fingernails and toenails concurrently at 2 and service for families enrolled in the IPCRR and has yielded a 4 years of age. Two additional features found commonly with wealth of information about PC characteristics and genotype-

E6 JAMA Dermatology Published online October 16, 2013 jamadermatology.com

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Pachyonychia Congenita in Pediatric Patients Original Investigation Research

affected infants, providing an initial clue. By 5 years of age, Table 3. Demographics and PC-Like Clinical Characteristics of 7 Participants With Mutations in the Connexin 30 Genea plantar keratoderma is seen in 75.3% of children in addition to the nail dystrophy and is often painful. Thus, the diagnosis Feature Data of PC should easily be suspected before kindergarten. Al- Demographic though genotyping should be performed through the registry Current age, y to subclassify the disease, the combination of data about age Mean (SD) 48.3 (30.5) at onset, concurrent dystrophy of fingernails and toenails, pal- Median (range) 37 (6-92) mar keratoderma, and the presence of oral leukokeratosis Age group, y and/or hoarseness, cysts, follicular keratoses, and natal teeth 0-5 0 helps the practitioner to suspect a specific PC subtype and 6-10 1 (14.3) counsel families, even before genotyping is completed. For ex- 11-17 0 ample, the presence of nail dystrophy at birth, especially in- ≥18 6 (85.7) volving all nails, predicts PC-K6a or PC-K17 (P < .001); the con- Sex comitant development of oral leukokeratosis and often Female 3 (42.9) hoarseness during the first year of life suggest the diagnosis Male 4 (57.1) of PC-K6a (P < .001); and the concurrence of natal teeth indi- Family history positive for PC 7 (100.0) cates PC-K17 (P < .001). In contrast, the development during Clinical childhood of palmar keratoderma, especially with later onset Dystrophy of other features, may signal PC-K16. The presence of an iso- Toenail 7 (100.0) lated dystrophic fingernail or toenail is quite uncommon in PC Fingernail 7 (100.0) (77 patients with fingernail involvement overall and 6 [5.5%] Keratoderma with only 1 fingernail involved; 106 with toenail involvement Plantar 3 (42.9) and 1 with only 1 toenail involved [0.9%]); the exception is PC- K6c, in which localized nail involvement is common.18,19 Palmar 3 (42.9) These features, although helping to differentiate among Pain the PC subtypes, also allow us to reject the old classifica- Plantar 2 (28.6) tions of PC-1 and PC-2 in pediatric PC. For example, cysts Palmar 2 (28.6) and natal teeth are most common in PC-K17 (formerly clas- Hearing loss 3 (42.9) sified as PC-2) but were both described in individuals with Abnormal hair 4 (57.1) PC-K6a (formerly PC-1) in our pediatric cohort and other Oral leukokeratoses 1 (14.3) published series.12,13 Hair disorders, at one time attributed Cysts 1 (14.3) to PC-2 KRT6B and KRT17 mutations, do not seem to be Hyperhidrosis 2 (28.6) associated with autosomal dominant PC more often than in Hypohidrosis 2 (28.6) the general population. Alopecia has recently been Learning disabilities 1 (14.3) described in association with severe PC manifestations in a Abbreviation: PC, pachyonychia congenita. patient with homozygous dominant missense mutations 20 a Unless otherwise indicated, data are expressed as number (percentage) of (from each affected parent) in keratin 17. These data sup- patients. port restructuring of the classification of PC from 2 different subtypes to a system that categorizes the disease based on phenotype correlations. By pooling data collected through the specific keratin mutations.12,16,21 IPCRR questionnaire and an auxiliary questionnaire, we were As a keratinopathy, PC is associated with increased cellu- able to address the early features of PC in an effort to aid early lar fragility and compensatory epidermal thickening at the sites diagnosis and bring attention to the profound effect this geno- of gene expression (particularly plantar keratoderma and nail dermatosis has on the quality of life of affected children and ado- dystrophy). The greater dystrophy of toenails vs fingernails and lescents. of the hallucal and fifth toenails likely reflect the propensity Our data confirm the frequent misdiagnosis of PC in pe- toward more pathological features with trauma. Natal teeth, diatric patients, showing that most children manifest the key which occurred in 13.9% of patients with PC overall and 86% features in the first year of life, but the diagnosis is made in of patients with PC-K17, have similarly been described in epi- only about 25%. This delay in diagnosis may lead to inappro- dermolysis bullosa simplex, which results from mutations in priate management (eg, topical or oral antifungals for pre- keratins 5 or 14.22 The role of keratin abnormalities in tooth for- sumed fungal infection, potent topical corticosteroids for pre- mation is poorly understood. sumed psoriasis) or incorrect information about prognosis and The genetic disorder that is most commonly confused inheritance (eg, mistaking hidrotic ectodermal dysplasia with with PC is hidrotic ectodermal dysplasia (Clouston syn- connexin 30 mutations for PC). The unifying feature of PC is drome), which results from mutations in connexin 30. Five nail dystrophy,17 although the clinical appearance of nails can of the 7 individuals from 2 families with Clouston syndrome be variable, even within families, and does not always show who were enrolled in the PC registry for genotyping also the classic V-shaped thickening. During the first year of life, showed multiple toenails and fingernails affected at birth, dystrophy of the fingernails and toenails occurs in most of the and several individuals developed painful plantar kerato-

jamadermatology.com JAMA Dermatology Published online October 16, 2013 E7

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Research Original Investigation Pachyonychia Congenita in Pediatric Patients

derma. However, the presence in hidrotic ectodermal dys- ager can improve the appearance of nails and keratoderma, plasia of hearing loss (a common feature of connexin gene thus improving patient coping. defects) and thin, sparse hair during childhood is not typical This study was limited by its retrospective nature, lead- of PC. Another genetic disorder with which PC could be con- ing to the risk of recall bias, particularly when inquiring about fused results from mutation of FZD6. The FZD6 gene disease characteristics at onset during the pediatric years given encodes frizzled 6, a Wnt-signaling pathway receptor that is that most of the patients were well into their adult years when localized to the nail matrix; autosomal recessive mutations they responded to the questionnaire. Nevertheless, the data in FZD6 lead to hypertrophic nail dystrophy from birth in this registry database show that PC can be diagnosed dur- without plantar or palmar keratoderma.23,24 ing early childhood based on the constellation of clinical fea- Our study emphasizes the negative effect of the nail and tures and genotyping. Genotyping is currently performed at skin changes of PC in preteenaged and adolescent patients. no cost on a research basis (ie, not by a Clinical Laboratory Im- Most patients reported embarrassment, teasing, an effect on provement Amendment–approved laboratory) by registering dating, and limitations in participation in physical activities with the IPCRR (www.pachyonychia.org). The many genotype- with classmates. Early diagnosis and discussion allow proac- phenotype and subtype-phenotype correlations allow for early tive management of psychosocial issues, including how a child classification, prediction of clinical features and their age at can comfortably inform peers about the disorder or how a teen- onset, and optimal management.

ARTICLE INFORMATION 3. Leachman SA, Kaspar RL, Fleckman P, et al. 15. McLean WH, Smith FJ, Cassidy AJ. Insights into Accepted for Publication: June 27, 2013. Clinical and pathological features of pachyonychia genotype-phenotype correlation in pachyonychia congenita. J Investig Dermatol Symp Proc. congenita from the human intermediate filament Published Online: October 16, 2013. 2005;10(1):3-17. mutation database. J Investig Dermatol Symp Proc. doi:10.1001/jamadermatol.2013.6448. 4. Bowden PE, Haley JL, Kansky A, Rothnagel JA, 2005;10(1):31-36. Author Contributions: Dr Paller had full access to Jones DO, Turner RJ. Mutation of a type II keratin 16. Wilson NJ, Leachman SA, Hansen CD, et al. A all the data in the study and takes responsibility for gene (K6a) in pachyonychia congenita. Nat Genet. large mutational study in pachyonychia congenita. the integrity of the data and the accuracy of the 1995;10(3):363-365. J Invest Dermatol. 2011;131(5):1018-1024. data analysis. Study concept and design: Kenner-Bell, Paller. 5. Cogulu O, Onay H, Aykut A, et al. Pachyonychia 17. Ward KM, Cook-Bolden FE, Christiano AM, Acquisition of data: Shah, Boen, Schwartz, Paller. congenita type 2, N92S mutation of keratin 17 gene. Celebi JT. Identification of a recurrent mutation in Analysis and interpretation of data: Shah, Boen, Eur J Pediatr. 2009;168(10):1269-1272. keratin 6a in a patient with overlapping clinical Rademaker, Paller. 6. Covello SP, Smith FJ, Sillevis Smitt JH, et al. features of pachyonychia congenita types 1 and 2. Drafting of the manuscript: Shah, Boen, Schwartz, Keratin 17 mutations cause either steatocystoma Clin Exp Dermatol. 2003;28(4):434-436. Paller. multiplex or pachyonychia congenita type 2. Br J 18. Wilson NJ, Messenger AG, Leachman SA, et al. Critical revision of the manuscript for important Dermatol. 1998;139(3):475-480. Keratin K6c mutations cause focal palmoplantar intellectual content: Shah, Boen, Kenner-Bell, 7. McLean WH, Rugg EL, Lunny DP, et al. Keratin 16 keratoderma. J Invest Dermatol. 2010;130(2): Rademaker, Paller. and keratin 17 mutations cause pachyonychia 425-429. Statistical analysis: Shah, Boen, Rademaker, Paller. congenita. Nat Genet. 1995;9(3):273-278. 19. Akasaka E, Nakano H, Nakano A, et al. Diffuse Administrative, technical, and material support: 8. Smith FJ, Jonkman MF, van Goor H, et al. A and focal palmoplantar keratoderma can be caused Boen, Paller. by a keratin 6c mutation. Br J Dermatol. Study supervision: Boen, Paller. mutation in human keratin K6b produces a phenocopy of the K17 disorder pachyonychia 2011;165(6):1290-1292. Conflict of Interest Disclosures: DrPallerisa congenita type 2. Hum Mol Genet. 20. Wilson NJ, Pérez ML, Vahlquist A, et al. member of the International Pachyonychia 1998;7(7):1143-1148. Homozygous dominant missense mutation in Congenita Consortium without financial keratin 17 leads to alopecia in addition to severe compensation. No other disclosures were reported. 9. Smith FJ, Liao H, Cassidy AJ, et al. The genetic basis of pachyonychia congenita. J Investig pachyonychia congenita. J Invest Dermatol. Funding/Support: The genotyping for the IPCRR is Dermatol Symp Proc. 2005;10(1):21-30. 2012;132(7):1921-1924. conducted under grant 315.811099 from the 10. Clementi M, Cardin de Stefani E, Dei Rossi C, 21. Spaunhurst KM, Hogendorf AM, Smith FJ, et al. Pachyonychia Congenita Project to the University Pachyonychia congenita patients with mutations in of Dundee McLean/Smith laboratory. Avventi V, Tenconi R. Pachyonychia congenita Jackson-Lawler type. Br J Dermatol. KRT6A have more extensive disease compared with Role of the Sponsor: The funders had no role in the 1986;114(3):367-370. patients who have mutations in KRT16. Br J design and conduct of the study; in the analysis, Dermatol. 2012;166(4):875-878. and interpretation of the data; or in the 11. Munro CS. Pachyonychia congenita. Br J Dermatol. 2001;144(5):929-930. 22. Liu HH, Chen CJ, Miles DA. Epidermolysis preparation, review, or approval of the manuscript. bullosa simplex. ASDC J Dent Child. The Pachyonychia Congenita Project assisted in the 12. Eliason MJ, Leachman SA, Feng BJ, Schwartz 1998;65(5):349-353. collection of data for the manuscript. ME, Hansen CD. A review of the clinical phenotype of 254 patients with genetically confirmed 23. Naz G, Pasternack SM, Perrin C, et al. FZD6 Additional Contributions: Genotyping was encoding the Wnt receptor frizzled 6 is mutated in performed in the laboratory of W. H. Irwin McLean, pachyonychia congenita. J Am Acad Dermatol. 2012;67(4):680-686. autosomal-recessive nail dysplasia. Br J Dermatol. DSc, FRSE, and Frances J. D. Smith, PhD, University 2012;166(5):1088-1094. of Dundee, Scotland. 13. Fu T, Leachman SA, Wilson NJ, Smith FJ, Schwartz ME, Tang JY. Genotype-phenotype 24. Perrin C, Langbein L, Schweizer J. Expression REFERENCES correlations among pachyonychia congenita of hair keratins in the adult nail unit. Br J Dermatol. 2004;151(2):362-371. 1. Jadassohn J, Lewandowsky P. Pachyonychia patients with K16 mutations. J Invest Dermatol. congenita: keratosis disseminata circumscripts 2011;131(5):1025-1028. (follicularis): tylomata: leukokeratosis linguae. 14. McLean WH, Hansen CD, Eliason MJ, Smith FJ. Ikonographia Dermatol. 1906:29-31. The phenotypic and molecular genetic features of 2. Irvine AD, McLean WH. Human keratin diseases. pachyonychia congenita. J Invest Dermatol. Br J Dermatol. 1999;140(5):815-828. 2011;131(5):1015-1017.

E8 JAMA Dermatology Published online October 16, 2013 jamadermatology.com

Downloaded From: http://archderm.jamanetwork.com/ by a University of Utah User on 10/28/2013 Clinical dermatology • Original article CED Clinical and Experimental Dermatology

Can skin disease cause neuropathic pain? A study in pachyonychia congenita

T. Wallis,1 C. D. Poole2 and B. Hoggart3 1University Hospitals Southampton, Southampton General Hospital, Southampton, Hampshire, UK; 2Department of Primary Care & Public Health, Cardiff University, Cardiff, Glamorgan, UK; and 3Pain Management Research, Solihull Hospital, Heartlands NHS Foundation Trust, Solihull, UK doi:10.1111/ced.12723

Summary Introduction. Pachyonychia congenita (PC) is a rare skin disorder caused by an autosomal dominant mutation in one of five genes encoding keratin (K6a, K6b, K6c, K16 or K17; each defining one PC subtype). Pain is a prominent symptom, but its severity and type are poorly characterized. Methods. In total, 35 genotyped US patients with PC consented to clinical assessment including the quality of life (QoL) questionnaire EQ-5D-3L, the Brief Pain Inventory (BPI) and painDETECT. Abbreviated quantitative sensory testing (QST) was also performed, and included mechanical detection threshold (MDT), mechanical pain threshold (MPT), wind-up pain ratio (WUR) and vibration detection threshold (VDT). Results. Significant pain in patients with PC was confirmed, as indicated by mean BPI severity and interference of 4.2 Æ 1.7 and 4.4 Æ 2.2, respectively, as well as QoL impairment, as indicated by mean EQ-5D index of 0.69 Æ 0.18. PD identified neuropathic pain in 62% of patients, the remainder being nociceptive. The pain- DETECT score was most significantly related to EQ-5D index (R2 = 0.26, P = 0.02). The K17 and K6a subtypes exhibited significantly worse QoL (0.584 and 0.613 respectively) than the K16 and K6b subtypes (P = 0.02). In QST analysis, abnormal pressure pain (assessed as MPT) was frequently observed, with more than half of patients with PC affected (54%), and 57% of patients with K17 also exhibiting abnormality in minimum touch threshold (assessed as MDT, P < 0.05). Very few patients were receiving analgesic therapy appropriate for neuropathic pain. Conclusion. Significant neuropathic pain was observed in PC, which warrants appropriate treatment. The health states observed in this sample are at a level that the average US citizen would forfeit one-third of their remaining lifespan to avoid.

Introduction KRT17 (K17).1–3 These mutations classify PC clinical subtypes as PC-K6a, PC-K6b, PC-K6c, PC-K16 and PC- Pachyonychia congenita (PC) is a rare skin disorder K17, respectively.4 There are currently 619 genetically caused by an autosomal dominant mutation in one of confirmed cases of PC worldwide.5 PC prevalence in at least five keratin genes, KRT6A (encodes K6a pro- western developed nations is 0.9 cases per million,6 tein), KRT6B (K6b), KRT6C (K6c), KRT16 (K16) and and extrapolation suggests a worldwide PC population 7 Correspondence: Dr Tim Wallis, University Hospitals Southampton, of 6500, consistent with previous estimates. Southampton General Hospital, Tremona Road, Southampton, Hampshire, The clinical phenotype of PC is a triad of nail dys- SO16 6YD, UK trophy, plantar keratoderma (Fig. 1) and plantar E-mail: [email protected] pain.4,8 PC keratoderma is exquisitely painful, espe- Conflict of interest: the authors declare that they have no conflicts of cially on weight-bearing areas. Other common clinical interest. manifestations include: cysts, follicular hyperkeratosis, Accepted for publication 11 March 2015 oral leucokeratosis and palmar keratoderma.

ª 2015 British Association of Dermatologists Clinical and Experimental Dermatology 1 Can skin disease cause neuropathic pain? T. Wallis et al.

(a) (b) Methods This observational study had full ethics approval from the institutional review board [WIRB Study number IPCRR (registry): 20040468, WIRB Study number for the pain study: 20111060, Western IRB, Olympia, WA, USA], and all recruited subjects provided informed written consent.

Selection criteria and participants

(c) (d) Study subjects had to meet all of the following inclusion criteria: (i) age 18–90 years inclusive, (ii) prior diagnosis of PC with conﬁrmed genotype and (iii) ability to understand and comply with the study as judged by the investigator (BH). Patients were ineligible if they met any of the following exclusion criteria: (i) lack of consent capacity, (ii) secondary skin infections of the feet, (iii) diabetes-related or other distal sensory neuropathy, (iv) NeP conditions caused by any other injury, (v) any unstable disease incompatible with the study objectives or (vi) any psychiatric disorder con- (a–d) Callus state of plantar kertoderma in pachyony- Figure 1 founding reliable information-gathering. chia congenita evident in a selection of research subjects with different keratin subtypes: (a) K16, (b) K17, (c) K6a and (d) K6b. In total, 35 adult patients (17 men, 18 women; mean Æ SD age 45.8 Æ 16.0, range 18–84 years) with genetically conﬁrmed PC registered in the Interna- Keratins are key structural proteins that impart tional PC Research Registry (IPCRR)12 and attending a structural strength and integrity to epithelial cells and patient support meeting focusing on pain in PC were tissues. Defective processing of the keratin cytoskeleton invited to participate in the study. The study comprised can cause fragility in epithelial cells and tissues in three different components: (i) self-assessment question- which the defective keratin is expressed.9 PC keratins naires [the Brief Pain Inventory ((BPI),13 a neuropathic are predominantly expressed in keratinocytes of the pain symptom inventory (painDETECT)14 and a generic nail, palmoplantar skin and oral mucosa. QoL instrument (EQ-5D-3L)15]; (ii) standardized quanti- Physical pain is either nociceptive or neuropathic. tative sensory testing (QST);16 and (iii) clinical evalua- Nociceptive pain arises from actual or threatened tis- tion by an experienced pain physician. The clinical sue damage, and involves the activation of nocicep- assessments were conducted within a fully air- tors, whereas neuropathic pain (NeP) arises from conditioned hotel, in Philadelphia (USA) during August damage or disease to the somatosensory nervous sys- 2010. Patients continued all their regular medications tem.10 Patients with NeP may experience abnormal including analgesics during the study period (Table 1). sensations such as burning, tingling or numbness, in addition to persistent or paroxysmal pain independent of painful stimuli. The treatment of NeP is often unre- Questionnaires sponsive to typical agents used for nociceptive pain.11 Brief Pain Inventory. The short-form BPI (BPI-SF)13 is Pain is an over-riding symptom of PC, but there has a validated questionnaire frequently used for measur- been little research into its nature. In this study, we ing pain, and is suitable for either patient self-report- investigated pain in PC, including its characteristics, ing or interviewer administration.17 BPI-SF records severity and effect on quality of life (QoL). Validated the severity of pain and its effect on daily function- patient-reported outcome questionnaires (PROs) and ing. Each BPI question is scored from 0 to 10, from clinical assessment measured the prevalence of neuro- which two indices, pain severity (ranging from ‘no pathic pain and QoL within PC subtypes. A better pain’ to ‘worst imaginable pain’) and pain interfer- understanding of the nature of pain in PC and the ence with daily life (from ‘does not interfere’ to ‘com- associated burden of illness should enable more appro- pletely interferes’), are derived. priate management.

2 Clinical and Experimental Dermatology ª 2015 British Association of Dermatologists Can skin disease cause neuropathic pain? T. Wallis et al.

Table 1 Subject characteristics by PC subtype.*

PC (keratin) subtype

Parameter K6a K6b K16 K17 Overall P n 11 3 14 7 35 Sex (M/F), % 55/46 67/33 64/36 14/86 51/49 0.16† Age, years 45 (8) 44 (23) 52 (19) 41 (17) 47 (16) 0.44‡ BMI, kg/m2 25.6 Æ 6.2 23.5 Æ 2.4 27.2 Æ 4.9 27.8 Æ 5.2 26.5 Æ 5.2 0.60‡ BPI severity 4.7 Æ 1.5 3.4 Æ 0.6 4 Æ 1.8 4.2 Æ 1.9 4.2 Æ 1.7 0.60‡ BPI interference 5.3 Æ 2.4 3.1 Æ 1.8 3.5 Æ 1.7 5.1 Æ 2.5 4.4 Æ 2.2 0.13‡ EQ-5D index 0.61 Æ 0.20 0.87 Æ 0.12 0.76 Æ 0.11 0.58 Æ 0.21 0.70 Æ 0.18 0.02‡ EQ-5D VAS 75.4 Æ 15.3 94 Æ 3.6 79.4 Æ 11.8 64.8 Æ 25.3 77 Æ 17 0.08‡ painDETECT score 19.1 Æ 6.5 10.7 Æ 10.3 12.2 Æ 6.4 17.3 Æ 7.3 15 Æ 7 0.07‡ painDETECT class, % Nociceptive 27 67 50 14 37 0.02† Mixed 9 0 36 71 31 Neuropathic 64 33 14 14 31 Clinical examination, % Nociceptive 9 67 57 43 40 0.33† Mixed 18 0 21 14 17 Neuropathic 73 33 21 43 43 QST, % MDT abnormal 9 0 14 57 20 0.05† MPT abnormal 55 0 64 57 54 0.25† WUR abnormal 27 33 29 29 29 0.10† VDT abnormal 9 0 7 29 11 0.44† QST abnormality, % None 18 67 14 14 20 0.18† 1 5533361437 2+ 27 0507143 Painkiller used, % Opiate 0 0 7 43 11 0.03† Anticonvulsant 0 0 0 14 3 0.25† Antidepressant 18 0 7 14 11 0.76† NSAID 55 0 71 71 57 0.12† Aspirin/paracetamol 18 07080.52† Topical analgesia 18 07080.52† Other analgesia 9 07050.82†

BPI, Brief Pain Inventory; MDT, mechanical detection threshold; MPT; mechanical pain threshold; NSAID, nonsteroidal anti-inflammatory drug; PC, pachyonychia congenita; QST, quantitative sensory testing; VAS, visual analogue score; VDT, vibration detection threshold; WUR; wind-up pain ratio. *Data are mean Æ SD unless otherwise specified; †v² test for proportions; ‡ANOVA. painDETECTTM. The painDETECT questionnaire is a patient responses to the EQ-5D index questionnaire validated patient-reported outcome (PRO) instrument were converted to a single summary societal prefer- with high sensitivity and specificity for predicting ence measure of health utility using the US tariff.19 NeP.14 Four sections assess the intensity, localization, The EQ-5D VAS is a linear score, with 100 represent- pattern and quality of pain. Summary scores ≤ 12 ing ‘best imaginable health state’ and 0 representing indicate that NeP is unlikely, scores of 13–18 suggest ‘worst imaginable health state’. pain of uncertain/mixed aetiology and scores ≥ 19 denote prominent NeP.18 Quantitative sensory testing EQ-5D. EQ-5D-3L assesses health-related (HR)QoL15 QST is a noninvasive method for testing the integrity and consists of a five-item questionnaire (mobility, self- of sensation in neurological disorders.16 We used a care, usual activities, pain/discomfort and anxiety/ four-stage paradigm to measure both loss of sensation depression) and a visual analogue scale (VAS). Each and hypersensitivity in the lower limbs. Time con- question has three levels of response (‘no problem’, straints limited these tests to those expected to have ‘some problems’ or ‘severe problems’). Individual the greatest sensitivity in this population, namely:

ª 2015 British Association of Dermatologists Clinical and Experimental Dermatology 3 Can skin disease cause neuropathic pain? T. Wallis et al.

(a) (b)

Figure 2 Individual patient quantitative sensory testing (QST) parameter scores by pachyonychia congenita (PC) subtype. (a) Mechani- cal detection threshold (MDT) (green, K6a; purple, K6b; blue, K16; orange K17). (b) Mechanical pain threshold (MPT) (solid, right foot; clear, left foot). (c) wind-up ratio (WUR). Z-scores represent individual score standardized for population reference values speciﬁc to sex, age and anatomical site. Scores within Æ 1.96 SD of the age-/sex-/site-speciﬁc population mean are considered to be ‘normal’ (grey shading), while scores > 1.96 SD represent hypersensitivity to evoked stimuli and scores < 1.96 SD represent loss of function. (d) Vibra- tion detection threshold (VDT).

(i) mechanical (touch) detection threshold (MDT); (ii) population mean reference values speciﬁc to sex, age pressure-evoked pain (mechanical pain threshold; group and anatomical site.16 MPT; (iii) repeated tapping of the skin using a non- painful ﬁlament (wind-up pain ratio; WUR); and (iv) Clinical examination vibration detection threshold (VDT). Thermal testing of patients with PC was considered inappropriate A clinical examination taking 30 min was carried out because of the dense plantar hyperkeratosis. Operators by an experienced consultant specializing in the man- were fully competent in the application of QST, and agement of chronic pain (BH). patients received practice and training prior to data collection. Statistical analysis The QST protocol was similar to that developed by Rolke et al.,16 except that only the feet (the primary Statistical comparisons between PC subtypes were site for PC pain) were assessed, in line with recom- made by ANOVA for continuous variables and the v² mended practice for bilateral NeP conditions. QST test for categorical variables. Correlations between scores were converted to z-scores using published continuous variables were made with curve estimation

4 Clinical and Experimental Dermatology ª 2015 British Association of Dermatologists Can skin disease cause neuropathic pain? T. Wallis et al.

regression. All analyses were conducted using IBM SPSS Statistics (v20; IBM Inc., Armonk, NY,USA).

Results

Subjects Subject numbers in each gene classiﬁcation were: K6A n = 11, K6B n = 3, K16 n = 14 and K17 n = 7, and mean body mass index (BMI) was 26.5 Æ 5.2 kg/m2. Comorbidity was generally low in this cohort of rela- tively young active adults. There were no statistically signiﬁcant demographic differences between PC subtypes (Table 1).

Pain characteristics BPI pain severity and pain interference were similar across subtypes, with overall means of 4.2 Æ 1.7 and 4.4 Æ 2.2, respectively. A trend for higher painDETECT scores was observed Relationship between painDETECT score and quality of Figure 3 in the K6a and K17 subtypes (mean scores of life (QoL) as represented by EQ-5D index (US tariff). painDETECT: final score ≤ 12 indicates that neuropathic pain is unlikely, 19.1 Æ 6.5 and 17.3 Æ 7.3 respectively) compared scores of 13–18 indicate pain of uncertain/mixed aetiology and with K6b or K16 subtypes (mean scores of 10.7 Æ and scores ≥ 19 indicate a significant neuropathic component to 10.3 and 12.2 Æ 6.4, respectively; P = 0.07). This the patient’s pain. EQ-5D index: US-specific societal preference for was reflected in a higher proportion of NeP cases in health states. A score of 1 indicates ‘perfect health’, i.e. no the K6a group (64%) (P = 0.02). QoL appeared to be impairments; a score of 0 indicates a state of health equivalent to death i.e. the ‘average’ US citizen would trade all their lowest in the K17 group, with mean EQ-5D index and remaining lifespan to avoid such health states. Best fit trend line VAS scores of 0.58 Æ 0.21 (P = 0.05) and 64.8 Æ represent cubic function (R2 = 0.26, P = 0.02), 25.3 (P < 0.08), respectively.

(a) (b)

Figure 4 Association of EQ-5D index (US tariff) with Brief Pain Inventory (BPI) indices: (a) BPI Severity (quadratic function R2 = 0.19, P = 0.121) and (b) BPI Interference (linear function R2 = 0.12, P < 0.05).

ª 2015 British Association of Dermatologists Clinical and Experimental Dermatology 5 Can skin disease cause neuropathic pain? T. Wallis et al.

(a) (b)

Figure 5 Association between quantitative sensory testing (QST) item abnormality and painDETECT score. (a) mechanical detection threshold (t-test P = 0.54); (b) mechanical pain threshold (t-test P = 0.23); (c) wind-up ratio (t-test P = 0.98); (d) vibration detection threshold (t-test P = 0.85). For each parameter, ‘1.00’ denotes presence of an abnormal score, i.e. patient has at least one z-score either > 1.96 SD or > À1.96 SD the age-/sex-/site-standardized population mean. ‘0.00’ denotes no abnormal score recorded.

QST indicated that most patients with PC experi- patients took medications considered helpful for enced abnormal detection of pressure-invoked pain; neuropathic pain (antidepressant 11%, anticonvul- 55% of K6a, 64% of K16, and 57% of K17 patients sant 3%). showed altered MPT function, whereas this was not the case for the K6b subtype. In addition, 57% of K17 Correlations patients exhibited abnormal detection threshold for touch (assessed by MDT) (P < 0.05). Variation in QST There was a signiﬁcant cubic correlation (P = 0.02) parameters is illustrated in Fig. 2. between painDETECT score and EQ-5D index (Fig. 3), The predominant (by 57%) type of pain relief used but not BPI severity (P = 0.121). A weak negative by these patients was nonsteroidal anti-inﬂammatory linear relationship between EQ-5D index and BPI pain drugs (NSAIDs), although opiates were used by interference (Fig. 4) was also observed. QST abnormal- nearly half of K17 patients (43%) (P = 0.03). Few ity was not discriminated by painDETECT score

6 Clinical and Experimental Dermatology ª 2015 British Association of Dermatologists Can skin disease cause neuropathic pain? T. Wallis et al.

K17 and K6a patients, which were the groups with the highest prevalence of NeP. Patients with PC often experience severe pain on walking. The majority of patients in our study had increased sensitivity in the feet on QST testing, suggesting that some of this discomfort is neuropathic in origin. This was particularly evident in the K6a and K17 patients. Time constraints and limited patient availability necessitated a reduced profile of QST testing. This might be a potential weakness of our study; nevertheless, others have shown that NeP can be detected prior to and following knee surgery using a similarly reduced profile of QST tests.24,25 The triad for diagnosing NeP includes clinical examination. In this study, the diagnosis of NeP by clinical Figure 6 painDETECT scores (PD_FINALSCORE) by clinical assessment classification (ANOVA P < 0.08). Diagnostic clinical examination correlated with the painDETECT question- classification (DX_CLINICAL_C): ‘No’, no clinical evidence of neu- naire, indicating that this instrument might prove use- ropathic pain; ‘Mixed’, mixed nociceptive and neuropathic pain; ful for clinicians less experienced in diagnosing NeP in ‘Yes’, predominantly neuropathic pain. this and other skin diseases.11

Conclusion (Fig. 5), although a nonsignificant trend (P < 0.08) was seen for variation in painDETECT scores across We have shown that NeP is an appreciable problem in clinical investigator assignment of nociceptive pain, patients with PC. Recognizing NeP is important, as it mixed pain and NeP (Fig. 6). is known to be associated with significant impairment of QoL and requires appropriate medication.26 The painDETECT questionnaire combined with clinical Discussion assessment will aid diagnosis of NeP in skin disease The nature of pain in PC (NeP or nociceptive pain) and improve patient care. and a quantitative assessment of its effect on QoL has not been reported previously. As NeP is poorly Acknowledgements recognized,20 understanding the varied aetiology of pain in PC will improve research to guide pain We thank the QST operators J. Clapham and R. Dug- management. All groups in this study reported mod- gan for their assistance in undertaking this study, and erate to severe pain as assessed by the BPI. Pain R. Knaggs for his help with reviewing the article. We interfered with daily life in all PC subgroups of PC are grateful to the Pachyonychia Congenita Project for and there was weak evidence to suggest that the funding, and to M. Schwartz (Director) for help with K17 and K6a groups may have been more severely the project. We are indebted to all the patients with affected, although this difference was not statistically PC who participated in the study; it was a privilege to significant (P = 0.13). work with them. One-third of patients had predominately NeP while another third had a mixture of NeP and nociceptive pain, suggesting a five-fold greater prevalence of NeP in What’s already known about this topic? patient with PC than in the general population.21,22 ● Despite this, only a minority of patients with PC were PC is a rare skin disorder caused by an autoso- receiving medications considered to be helpful in allevi- mal dominant mutation in keratin-encoding ating NeP. genes. ● The mean EQ-5D index observed in this study Pain is a significant symptom, and negatively (0.67) is equivalent to that of patients with cardiac ill- affects QoL. ness.23 Our analysis showed the poorest QoL in the

ª 2015 British Association of Dermatologists Clinical and Experimental Dermatology 7 Can skin disease cause neuropathic pain? T. Wallis et al.

11 Haanpa€a€ M, Attal N, Backonja M et al. NeuPSIG guidelines on neuropathic pain assessment. Pain 2011; What does this study add? 152:14–27. 12 The Pachyonchia Congenita Project. Available at: http: ● Pain is a primary symptom of PC, and a signifi- //www.pachyonychia.org/. cant proportion of this paitn is neuropathic in 13 Cleeland C. Pain assessment in cancer. In: Effect of Cancer nature. on Quality of Life (Osoba D ed). Boca Raston, FL: CRC ● The effect of PC on QoL in adults has been Press Inc., 1991; 293–305. quantified. 14 Freynhagen R, Baron R, Gockel U, Tolle€ TR. ● Understanding the aetiology of pain in PC will painDETECT: a new screening questionnaire to identify enable more appropriate treatment. neuropathic components in patients with back pain. Curr Med Res Opin 2006; 22: 1911–20. 15 The EuroQoL Group. EuroQol–a new facility for the measurement of health-related quality of life. Health Policy 1990; 16: 199–208. References 16 Rolke R, Baron R, Maier C et al. Quantitative sensory 1 Bowden PE, Haley JL, Kansky A et al. Mutation of a type testing in the German Research Network on Neuropathic II keratin gene (K6a) in pachyonychia congenita. Nat Pain (DFNS): standardized protocol and reference values. 123 – Genet 1995; 10: 363–5. Pain 2006; : 231 43. 2 McLean WH, Rugg EL, Lunny DP et al. Keratin 16 and 17 Keller S, Bann CM, Dodd SL et al. Validity of the Brief Pain keratin 17 mutations cause pachyonychia congenita. Nat Inventory for use in documenting the outcomes of patients 20 – Genet 1995; 9: 273–8. with noncancer pain. Clin J Pain 2004; : 309 18. 3 Smith F. A mutation in human keratin K6b produces a 18 Bennett MI, Attal N, Backonja MM et al. Using screening 127 – phenocopy of the K17 disorder pachyonychia congenita tools to identify neuropathic pain. Pain 2007; : 199 type 2. Hum Mol Genet 1998; 7: 1143–8. 203. 4 McLean WHI, Hansen CD, Eliason MJ, Smith FJD. The 19 Shaw JW, Johnson JA, Coons SJ. US valuation of the EQ- phenotypic and molecular genetic features of 5D health states: development and testing of the D1 43 – pachyonychia congenita. J Invest Dermatol 2011; 131: valuation model. Med Care 2005; : 203 20. 1015–17. 20 Baron R, Binder A, Wasner G. Neuropathic pain: 5 International PC Research Registry. Worldwide locations diagnosis, pathophysiological mechanisms, and 9 – of PC PATIENts in the IPCRR, 2015. Available at: http:// treatment. Lancet Neurol 2010; : 807 19. www.pachyonychia.org/upload/GRAPHS/ 21 Torrance N, Smith BH, Bennett MI, Lee AJ. The IPCRRLocationPCers_1.pdf (accessed 7 March, 2015). epidemiology of chronic pain of predominantly 6 Wikipedia. List of countries and dependencies by neuropathic origin. Results from a general population 7 – population. Available at: http://en.wikipedia.org/wiki/ survey. J Pain 2006; : 281 9. List_of_countries_and_dependencies_by_population 22 Bouhassira D, Lanteri-Minet M, Attal N et al. Prevalence (accessed 7 March, 2015). of chronic pain with neuropathic characteristics in the 136 – 7 Kaspar RL. Challenges in developing therapies for rare general population. Pain 2008; : 380 7. diseases including pachyonychia congenita. J Investig 23 Sullivan PW, Ghushchyan V. Preference-based EQ-5D Dermatol Symp Proc 2005; 10:62–6. index scores for chronic conditions in the United States. 26 – 8 Eliason MJ, Leachman SA, Feng B, Schwartz ME, Hansen Med Decis Making 2009; : 410 20. CD. A review of the clinical phenotype of 254 patients 24 Martinez V, Fletcher D, Bouhassira D et al. The evolution with genetically confirmed pachyonychia congenita. JAm of primary hyperalgesia in orthopedic surgery: Acad Dermatol 2012; 67: 680–6. quantitative sensory testing and clinical evaluation 9 McLean WHI, Smith FJD, Cassidy AJ. Insights into before and after total knee arthroplasty. Anesth Analg 105 – genotype–phenotype correlation in pachyonychia 2007; : 815 21. congenita from the human intermediate filament 25 Wylde V, Palmer S, Learmonth ID, Dieppe P. The mutation database. J Investig Dermatol Symp Proc 2005; association between pre-operative pain sensitisation and 10:31–6. chronic pain after knee replacement: an exploratory 21 – 10 Treede R-D, Jensen TS, Campbell JN et al. Neuropathic study. Osteoarthritis Cartilage 2013; : 1253 6. pain: redefinition and a grading system for clinical and 26 Freynhagen R, Bennett MI. Diagnosis and management 339 – research purposes. Neurology 2008; 70: 1630–5. of neuropathic pain. Br Med J 2009; : 391 5.

8 Clinical and Experimental Dermatology ª 2015 British Association of Dermatologists ORIGINAL ARTICLE

A Large Mutational Study in Pachyonychia Congenita Neil J. Wilson1, Sancy A. Leachman2, C. David Hansen2, Alexandra C. McMullan1, Leonard M. Milstone3, Mary E. Schwartz4, W.H. Irwin McLean1, Peter R. Hull5 and Frances J.D. Smith1

Pachyonychia congenita (PC) is a rare autosomal dominant skin disorder characterized predominantly by nail dystrophy and painful palmoplantar keratoderma. Additional clinical features include oral leukokeratosis, follicular keratosis, and cysts (steatocysts and pilosebaceous cysts). PC is due to heterozygous mutations in one of four keratin genes, namely, KRT6A, KRT6B, KRT16,orKRT17. Here, we report genetic analysis of 90 new families with PC in which we identified mutations in KRT6A, KRT6B, KRT16,orKRT17, thereby confirming their clinical diagnosis. A total of 21 previously unreported and 22 known mutations were found. Approximately half of the kindreds had mutations in KRT6A (52%), 28% had mutations in KRT16, 17% in KRT17, and 3% of families had mutations in KRT6B. Most of the mutations were heterozygous missense or small in-frame insertion/ deletion mutations occurring within one of the helix boundary motif regions of the keratin polypeptide. More unusual mutations included heterozygous splice site mutations, nonsense mutations, and a 1-bp insertion mutation, leading to a frameshift and premature termination codon. This study, together with previously reported mutations, identifies mutation hotspot codons that may be useful in the development of personalized medicine for PC. Journal of Investigative Dermatology advance online publication, 17 February 2011; doi:10.1038/jid.2011.20

INTRODUCTION (Jadassohn and Lewandowski, 1906; Jackson and Lawler, Pachyonychia congenita (PC) is a rare genetic skin disorder 1951). A combination of factors have led to the suggestion that is associated with mutations in one of four keratin genes, that PC should be reclassified. First, the advent of molecular KRT6A, KRT6B, KRT16 or KRT17 (Bowden et al., 1995; genetics and the identification of the genes causing PC McLean et al., 1995; Smith et al., 1998). The most striking provide a rational means of classifying patients. Second, feature of PC is the painful and debilitating plantar clinical analysis of the large case series collected by the keratoderma (Figure 1). The mechanism underlying the International Pachyonychia Congenita Research Registry plantar pain is poorly understood; however, the formation (IPCRR), fully linked to molecular genetic data, has shown of blisters beneath the keratoderma is likely to be a major that there is considerable phenotypic overlap between the contributing factor. Palmar keratoderma is less frequent. Nail historical PC-1 and PC-2 subtypes (Eliason et al., 2011). Thus, dystrophy presents in variable forms, from very minor or a new molecular genetic classification has been adopted, almost absent nail changes through to the classic hyper- fully supported by members of the International Pachyony- trophic nail dystrophy that gives the condition its name chia Congenita Consortium, whereby the subtypes of PC refer (Leachman et al., 2005; Figure 1). Other epithelial structures to the mutated keratin gene (PC-6a for a patient carrying a can be affected, particularly the mucosae and the piloseba- K6a mutation, PC-6b, PC-16, PC-17, and so on). The cesous unit. A more detailed description of the disorder is designation ‘PC-U’ is used for cases where the causative given in McLean et al. (2011). gene is unknown (McLean et al., 2011). The new classifica- Historically, PC has been subdivided into two subtypes, tion will (a) help discourage publication of spurious case PC-1 (Jadassohn–Lewandowski type) or PC-2 (Jackson–Lawler reports lacking molecular data (often with misdiagnosed type), on the basis of the clinical presentation alone cases or cases with coincidental findings unrelated to PC) and (b) allow better prognostic predictions and patient counseling, especially when referenced to the IPCRR clinical 1 Division of Molecular Medicine, University of Dundee, Dundee, UK; data set. 2Department of Dermatology, University of Utah, Salt Lake City, Utah, USA; 3Department of Dermatology, Yale University, New Haven, Connecticut, The 54 human keratins belong to the intermediate filament USA; 4PC Project, Salt Lake City, Utah, USA and 5Department of protein family that consists of at least six types; keratins make Dermatology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada up the type I and type II intermediate filament proteins. A Correspondence: Frances J.D. Smith, Division of Molecular Medicine, major function of keratins is to form structural cytoskeletal Medical Sciences Institute, University of Dundee, Dundee DD1 5EH, UK. networks within epithelial cells that allow cells to withstand E-mail: [email protected] everyday stress and physical trauma. Keratins are expressed Abbreviations: IPCRR, International Pachyonychia Congenita Research Registry; K, keratin protein; KRT, keratin gene; PC, pachyonychia congenita in pairs in tissue-specific and differentiation-specific Received 26 August 2010; revised 29 December 2010; accepted 5 January patterns (Lane, 1993). The keratins associated with PC, 2011 K6a, and K16, K6b and K17 are predominantly expressed

& 2011 The Society for Investigative Dermatology www.jidonline.org 1 NJ Wilson et al. Pachyonychia Congenita

K6a p.Asn171Lys (siblings) K6a p.Asn171Lys (siblings)

K6a p.Asn171Ser

K6a p.Asn172del K6a p.Leu468Pro

K16 p.Asn125Ser K16 p.Arg127Pro K16 p.Arg127Pro K17 p.Asn92_Leu99del

Figure 1. Clinical features of pachyonychia congenita (PC). Plantar keratoderma and nail dystrophy in PC patients with known keratin mutations. Mutations are annotated. Note the variation in severity and clinical appearance of these features.

in basal/suprabasal layers of palmoplantar skin, as well as in The IPCRR was established in 2004 by the patient epidermal appendages and oral mucosa. In PC, these advocacy group, Pachyonychia Congenita Project (http:// epithelial cell compartments are rendered fragile by the www.pachyonychia.org). At the time of writing, 478 families expression of dominant-negative mutant keratins. All cases of are registered (928 individuals), 223 families have completed PC with a confirmed molecular diagnosis, including those in the detailed questionnaire (Eliason et al., 2011) and 199 the literature, http://www.interfil.org (Szeverenyi et al., families have undergone genetic testing. Genetic testing 2008), and those in this study, are due to heterozygous results from some of these cases have been previously dominant-negative mutations, inherited as an autosomal published (McLean et al., 1995; Smith et al., 1997, 2000, dominant trait. Although there are a small number of case 2005; Liao et al., 2007a; Oh Adib et al., 2008; Cogulu et al., reports of PC with recessive inheritance in the literature, there 2009; Gruber et al., 2009). are no reports of recessive PC with a confirmed molecular Here, we present the findings of 90 new families with diagnosis. mutations in KRT6A, KRT6B, KRT16,orKRT17. Within this All keratins share a common protein structure consisting of case series, we identified 21 previously unreported mutations a central a-helical rod domain of 310 amino acids subdivided (22 families) and 22 known mutations (68 families). This into the 1A, 1B, 2A, and 2B domains. These domains are mutation analysis study not only confirms the clinical connected by non-helical linker regions, L1, L12, and L2. The diagnosis of these individuals but, together with previously rod domain is flanked by short regions of sequence homology reported mutations (http://www.interfil.org), also identifies (H1 and H2 regions), followed by the variable, non-helical mutation hotspot codons that may be useful in the develop- head (V1) and tail domains (V2). At either end of the rod ment of future allele-specific therapies. domain are the helix boundary motifs (the helix initiation motif and the helix termination motif). These highly RESULTS conserved motifs are thought to be important in mediating Clinical details end-to-end interactions during filament assembly. The All individuals involved in this study were recruited through majority of mutations in PC occur in these helix boundary the IPCRR, an ongoing research program to identify PC motifs, emphasizing the critical importance of these se- patients worldwide. This research registry is approved by an quences for correct keratin filament formation and the institutional review board that complies with all principles of mechanical resilience of epithelial cells. the Helsinki Accord (Western IRB Study no. 20040468). An

2 Journal of Investigative Dermatology NJ Wilson et al. Pachyonychia Congenita

important part of this study was the detailed clinical Both common and rare dominant mutations cause PC consultations that were performed for all cases analyzed. Within this large PC case series, pathogenic mutations were This bank of data allows us now, and in the future as the identified in the KRT6A gene in approximately half (52%) of number of cases analyzed increases, to identify any useful the kindreds, whereas 28% had mutations in KRT16, 17% genotype–phenotype correlation for PC. The predominant had defects in KRT17, and 3% had mutations in KRT6B (see clinical features of individuals involved in this study are Figure 2 and Supplementary Table S1 online). Mutations in summarized in Supplementary Table S1 online. Of the 90 KRT6A also account for B50% of previously reported cases families analyzed, 36 represent familial occurrence of PC, of PC (http://www.interfil.org), consistent with our finding with many showing autosomal dominant inheritance here that this is the predominant PC gene. The majority of the through several generations; the remaining 54 cases represent mutations we identified in all four genes were heterozygous spontaneous mutations. missense mutations occurring within one of the helix KRT6B boundary motif regions. In addition, we found some 3% small in-frame insertion/deletion mutations and, in particular, the common K6a p.N172del mutation was identified in 16 families.

An unusual V2 domain mutation in one PC family A more unusual mutation identified was a 1-bp insertion KRT17 17% in exon 9 of KRT6A, the last exon of this gene (K6a c.1511_1512insG). This insertion results in a frameshift and KRT6A a premature stop codon just two amino acids upstream of the KRT6A KRT16 52% natural stop codon, whereby the last 60 amino acids of the KRT17 K6a V2 domain are exchanged for a foreign peptide of 58 KRT16 amino acids. Protein–protein BLAST (basic local alignment KRT6B 28% search tool) analysis showed that this mutant peptide sequence has no significant similarity to any human protein (data not shown; http://blast.ncbi.nlm.nih.gov/). Kyte–Doolit- tle hydrophilicity analysis revealed that the normal K6a V2 domain consists of alternating hydrophobic and hydrophilic sequences, followed by a short hydrophilic C terminus (Figure 3), consistent with the glycine-loop structure Figure 2. Mutational spectrum in pachyonychia congenita (PC). Pie chart proposed by Steinert for keratin variable domains (Korge showing percentage of families in this study with mutations in the four keratin et al., 1992). In contrast, the mutant V2 domain is almost genes, namely, KRT6A, KRT6B, KRT16, and KRT17. completely hydrophilic (Figure 3). In terms of protein

a Normal K6a V2 domain b Frameshift mutant K6a V2 domain Kyte–Doolittle hydrophilicity: Window = 7 Kyte–Doolittle hydrophilicity: Window = 7 4.0 4.0 3.0 3.0 2.0 2.0 1.0 1.0 0.0 0.0 –1.0 –2.0 –1.0 Hydrophilicity –3.0 –2.0 –4.0 10 20 30 40 50 10 20 30 40 50

Robson–Garnier 2° structure: Window = 7 Robson–Garnier 2° structure: Window = 7

Helix Sheet Turn Helix Sheet Turn

400 600 200 400 0 200 –200 0 –400 –200 Robson–Garnier –600 –400 10 20 30 40 50 10 20 30 40 50

Figure 3. Kyte–Doolittle hydrophilicity analysis of normal and mutant K6a V2 domain. (a) The normal K6a V2 domain consists of alternating hydrophobic and hydrophilic sequences, followed by a short hydrophilic C terminus, whereas (b) the mutant V2 domain is almost completely hydrophilic.

www.jidonline.org 3 NJ Wilson et al. Pachyonychia Congenita

secondary structure predicted by Robson–Garnier analysis, Nonsense mutations in a few PC families the normal K6a V2 domain is predicted to adopt three large In addition, two heterozygous nonsense mutations were also areas of sheet conformations, separated by short regions identified in PC cases. One of these was identified in the 2B predicted to adopt turn conformations (Figure 3). This domain of K6a, p.Gln435X, which is predicted to lead to contrasts with the mutant V2 domain, which is predicted to expression of a truncated dominant-negative K6a protein, consist largely of turn conformation with one helical region lacking the end of the rod domain and the tail domain. near the C terminus (Figure 3). This in silico analysis Because this mutation is close to the natural stop codon of underscores the fact that the mutant polypeptide is very K6a, it is likely to escape nonsense-mediated decay to some different in both sequence and predicted secondary structure extent and, therefore, be expressed as mutant polypeptide from the wild-type K6a tail domain, consistent with a (Frischmeyer and Dietz, 1999). Analogous mutations have dominant-negative gain-of-function mutation, as seen in been seen in K5 in dominantly inherited epidermolysis other keratinizing disorders due to C-terminal frameshift bullosa simplex (Muller et al., 1999; Livingston et al., mutations in K1 (Sprecher et al., 2001, 2003; Richardson et al., 2001). The other nonsense mutation was found within the 2006) or K5 (Sprecher et al., 2003). In the case of loricrin head domain of K16, p.Lys15X. Analogous premature keratoderma, a similar C-terminal gain-of-function mutation has termination codon mutations just downstream of the ATG been shown to lead to creation of a new nuclear localization codon have been reported in other dominant keratin signal, which in turn leads to nuclear accumulation of mutant disorders, including K5 in Dowling–Degos disease (Betz protein (Ishida-Yamamoto et al., 2000). The mutant K6a et al., 2006; Liao et al., 2007b) and in K14 in Naegeli polypeptide sequence generated here was analyzed for syndrome (Lugassy et al., 2006). Although it remains some- potential nuclear localization signal motifs (http://cubic.bioc. what unclear whether these mutations act via haploinsuffi- columbia.edu/services/predictNLS/), but none were found. ciency or via expression of a dominant-negative mutant protein through use of an alternative initiation codon Splice site mutations identified in five kindreds (McLean et al., 2003), it is however clear that they exhibit Interestingly we also detected four previously unreported dominant inheritance (Betz et al., 2006; Lugassy et al., 2006). splice site mutations (in five families) at the intron 1/exon 2 Unfortunately, in the case of the two nonsense mutations boundary of KRT6A. These are clearly inherited in an identified here, it was not possible to obtain tissue to allow autosomal dominant manner. Unfortunately, we were unable analysis of mRNA or protein. to obtain mRNA from lesional skin that would allow analysis of the effects of these genomic mutations on RNA splicing. A spectrum of keratin mutations cause PC One possible predicted consequence of this type of mutation For each of the four genes associated with PC, it has been is skipping of exon 2; however, as this is an out-of-frame suggested that there are some codons that represent mutation exon, its deletion would lead to a frameshift and premature hotspots, as well as several rare or even family-specific muta- termination codon. This is unlikely to create a dominant- tions. Our results confirm the previously identified mutation negative mutant protein, because nonsense-mediated mRNA hotspots and also identify 21 previously unreported mutations. decay is predicted to occur, leading to loss of expression of Table 1 summarizes the data from previous publications this allele. In the case of KRT5 and KRT1, both of which are (http://www.interfil.org) together with the data from this large type II keratin genes closely related to KRT6A, analogous case series. mutations have been reported affecting the intron 1 splice sites (Rugg et al., 1999; Terron-Kwiatkowski et al., 2002). In DISCUSSION both these genes, the mutation led to activation of an The mutation results from this PC case series of 90 families, identical cryptic splice site in exon 1, producing a 66- together with those from at least 131 previously reported nucleotide (22 amino acid) in-frame deletion. Given the cases (http://www.interfil.org), provides a large data set for strong sequence homology between these genes, it is analysis in terms of where mutations occur within the keratin probable that a similar mechanism will also occur with these protein, the most common mutations, the types of mutations KRT6A splice site mutations. found, and allows for preliminary genotype–phenotype

Table 1. Summary of mutations identified in this study and previous publications Number of Number of Gene different mutations recurrent mutations Most common mutation site

KRT6A 39 13 K6a p.Asn171—as missense or deletion mutation (K6a p.Asn172del), this codon is mutated in 46% of those with KRT6A mutations, of which p.Asn172del accounts for 30% of all KRT6A mutations KRT16 19 8 K16 p.Leu132Pro in 23% of families with KRT16 mutations KRT17 22 8 K17 p.Asn92Ser in 36% of families with KRT17 mutations KRT6B 4 1 K6b p.Glu472Lys in 71% of families with KRT6B mutations

4 Journal of Investigative Dermatology NJ Wilson et al. Pachyonychia Congenita

correlation. All cases with confirmed PC have a mutation in insertion–deletion mutations. Less common types identified one of the four keratin genes associated with this disease, include splice site and premature termination codon muta- KRT6A, KRT6B, KRT16, and KRT17. In the course of running tions. There are a number of mutation ‘hotspot’ codons for the international mutation screening service for PC, in each of the keratin genes associated with PC as well as concert with the IPCRR, we have received a small number mutations that appear to be family specific (Table 1). The of samples from isolated cases or families in which no most commonly mutated codon is K6a p.Asn171, either as a mutation was found in these four keratin genes. In these missense mutation (e.g. p.Asn171Lys and p.Asn171Ser) or as cases, careful review of the clinical phenotype by the a deletion mutation (designated as p.Asn172del using the International Pachyonychia Congenita Consortium clinicians Human Genome Variation Society guidelines; http://www. often has often led to correction of the diagnosis, confirmed hgvs.org). Approximately half of the families with identified by analysis of other keratin genes or non-keratin genes. For KRT6A mutations have a mutation at this site; the most example, a few cases, in which there is alopecia in addition common PC mutation is the p.Asn172del mutation that, to to nail dystrophy, have turned out to carry heterozygous date, has been found in 32 out of 221 PC families (14%) with connexin-30 mutations (Lamartine et al., 2000; Smith et al., known mutations (http://www.interfil.org and this study). 2002; van Steensel et al., 2003). Thus, Clouston syndrome Despite the increased number of genotyped individuals should be considered in the differential diagnosis for PC. and families, there is no apparent correlation between the Similarly, a few families presenting with painful but very clinical features observed and the protein domain harboring limited, circumscribed focal plantar keratoderma, with the mutation in PC. Nevertheless, there are two mutations in minimal or absent nail changes, have recently been shown KRT16 in which the actual amino acid substitution appears to to have mutations in the gene encoding K6c (KRT6C; Wilson correlate with the severity of the clinical phenotype (see Fu et al., 2010). Taking these families into account, we have a et al., 2011). There are also some differences in clinical small number of families (o5%) in which careful clinical features depending on the gene involved, such as the evaluation is consistent with a diagnosis of PC and in whom presence of natal teeth in many, but not all, individuals with we cannot detect a mutation in any exon or splice sites of the KRT17 mutations (Eliason et al., 2011). In PC, there can also four PC keratin genes. Genetic linkage analysis in at least one be variation in clinical severity between mutations in the such family has yielded statistically significant linkage to the same gene and even between individuals with the same vicinity of a keratin gene cluster (FJD Smith, unpublished mutation. Polymorphisms, copy number variation, environ- data), and so we conclude that there are either intronic or mental factors, lifestyle, and patient care may account for genomic deletion/rearrangement mutations that are missed some of this clinical variation. Therefore, an important by conventional PCR analysis, or that at least one other conclusion of this large study of a keratin disorder is that keratin gene or a nearby related gene can lead to a PC-like PC can really be considered as a spectrum of phenotypes phenotype in a minority of cases. Sequence analysis of these ranging from very mild to more severe, in which (a) the cases is ongoing in the laboratory. particular gene involved appears to have a moderate All PC causative mutations found to date are heterozygous influence on phenotype and (b) the specific mutation changes that exhibit autosomal dominant inheritance with a generally appears to have little influence on phenotype. proven, or very probable, dominant-negative pathomechan- The detailed clinical information that is obtained by the ism. It is important to note that no recessive cases of PC have IPCRR also identified several cases, which were not included been confirmed at the molecular level, despite a few case in this study, that presented with both typical and atypical reports appearing in the literature, in which, for example, features of PC and that were unlikely due to a mutation in any recessive inheritance may have been suggested by coin- of the four keratin genes, KRT6A, KRT6B, KRT16,orKRT17. cidental consanguinity. It is therefore important that case These were analyzed for mutations in other candidate genes. reports of already characterized genetic diseases be backed For example, several families presented with varying degrees up by molecular analysis, otherwise the literature may of alopecia in addition to some features typical of PC, which become misleading. The recurrence risk of a sporadic case was suggestive of Clouston syndrome (Lamartine et al., 2000; of a dominant disorder is very low (involving only the risk of Smith et al., 2002; van Steensel et al., 2003) and mutations gonadal mosaicism) which is difficult to estimate with were subsequently identified in GJB6 (which encodes con- certainty. In the dominant disorder achondroplasia OMIM nexin 30). Another candidate gene considered for individuals no. 100800, this is of the order of one in a few hundred presenting with palmoplantar keratoderma was the third K6 (Mettler and Fraser, 2000); however, in epidermolysis bullosa gene (KRT6C). We have identified mutations in KRT6C in simplex, this has been estimated at 2–5% (Pfendner et al., several families presenting with palmoplantar keratoderma, 2005). In contrast, the recurrence risk for a recessive condition but with only mild/no nail changes (Wilson et al., 2010). is as high as 25%. Thus, there are important genetic counseling Genetic analysis of individuals with PC not only confirms implications in the correct assignation of inheritance patterns. their clinical diagnosis but also aids in genetic counseling. So far, we have confirmed gonadal mosaicism in only one PC The identification of mutations is especially important for the family out of 199 analyzed from the IPCRR. design of future mutation-specific and/or gene-specific The majority of the mutations causing PC are located in therapies and, hence, the large, well-phenotyped, and fully one of the helix boundary motifs of the mutated protein, and genotyped case series we report here is an invaluable most of the causative variants are missense or small in-frame resource for future clinical trials.

www.jidonline.org 5 NJ Wilson et al. Pachyonychia Congenita

MATERIALS AND METHODS SUPPLEMENTARY MATERIAL Clinical material Supplementary material is linked to the online version of the paper at http:// Genomic DNA was obtained with informed consent and ethical www.nature.com/jid approval by an institutional review board that complies with all principles of the Helsinki Accord (Western IRB Study no. 20040468). REFERENCES Genomic DNA was extracted from peripheral blood lymphocytes Betz RC, Planko L, Eigelshoven S et al. (2006) Loss-of-function mutations in using standard procedures or from saliva collected in an Oragene the keratin 5 gene lead to Dowling-Degos disease. Am J Hum Genet DNA sample collection kit (DNA Genotek, Ontario, Canada) and 78:510–9 extracted according to the manufacturer’s protocol. Bowden PE, Haley JL, Kansky A et al. (1995) Mutation of a type II keratin gene (K6a) in pachyonychia congenita. Nat Genet 10:363–5 Mutation detection Cogulu O, Onay H, Aykut A et al. (2009) Pachyonychia congenita type 2, N92S mutation of keratin 17 gene: clinical features, mutation analysis The coding regions of KRT6A, KRT6B, KRT16, and KRT17 were and pathological view. Eur J Pediatr 168:1269–72 amplified using primers specific to the respective functional genes to Eliason MJ, Leachman SA, Feng BJ et al. (2011) A review of the clinical avoid amplification of KRT6C or pseudogenes (Supplementary Table phenotype in 254 patients with genetically-confirmed pachyonychia S2 online). All primers were checked for single-nucleotide poly- congenita. JAAD (unpublished) morphisms using Diagnostic SNPCheck (http://www.ngrl.org.uk/ Frischmeyer PA, Dietz HC (1999) Nonsense-mediated mRNA decay in health Manchester) and some were modified from our previous publica- and disease. Hum Mol Genet 8:1893–900 tions (Smith et al., 2005) to increase specificity. For each gene, there Fu T, Leachman SA, Wilson NJ et al. (2011) Genotype-phenotype correlations are two primer sets for the mutation hotspot exons to overcome the among pachyonychia congenita patients with K16 mutations. J Invest Dermatol (this issue) potential problem of very rare or as yet unidentified single- Gruber R, Wilson NJ, Smith FJ et al. (2009) Increased pachyonychia congenita nucleotide polymorphisms in primers designed to amplify these severity in patients with concurrent keratin and filaggrin mutations. regions. Previously unreported mutations were excluded from at Br J Dermatol 161:1391–5 least 90 control DNA samples (180 chromosomes) by sequencing or Ishida-Yamamoto A, Kato H, Kiyama H et al. (2000) Mutant loricrin is not restriction enzyme digests. For full-length genomic PCR reactions, crosslinked into the cornified cell envelope but is translocated into the Takara buffer and LA Taq polymerase (Lonza Biologics PLC, Slough, nucleus in loricrin keratoderma. J Invest Dermatol 115:1088–94 UK) were used and for the smaller PCR reactions and HotStarTaq Jackson ADM, Lawler SD (1951) Pachyonychia congenita: a report of six DNA Polymerase and buffer system (Qiagen, Crawley, UK) were cases in one family with a note on linkage data. Ann Eugen 16:142–6 used according to the manufacturer’s instructions. Specific PCR Jadassohn J, Lewandowski F (1906) [Pachyonychia congenita: keratosis conditions for each primer set are available on request. PCR disseminata circumscripta (follicularis). Tylomata. Leukokeratosis linguae.]. Vol. 1. Berlin: Urban and Schwarzenberg, 29 products were purified using QiaQuick PCR spin columns (Qiagen) Korge BP, Gan S-Q, McBride OW et al. (1992) Extensive size polymorphism or ExoSAP (using Exonuclease 1 and Shrimp Antartic Phosphatase, of the human keratin 10 chain resides in the C-terminal V2 subdomain New England Biolabs, Hitchin, Herts, UK), and sequenced using due to variable numbers and sizes of glycine loops. Proc Natl Acad Sci internal primers on an ABI 3100 Automated DNA sequencing USA 89:910–4 machine (Applied Biosystems, Foster City, CA) according to the Lamartine J, Munhoz Essenfelder G, Kibar Z et al. (2000) Mutations in GJB6 manufacturer’s instructions. cause hidrotic ectodermal dysplasia. Nat Genet 26:142–4 Lane EB (1993) Keratins. In: Connective Tissue and its Heritable Disorders. Protein bioinformatics Molecular, Genetic and Medical Aspects. (Royce PM, Steinmann B, eds). New York: Wiley-Liss, 237–47 Hydrophobicity (Kyte–Doolittle analysis) and protein secondary structure predictions (Robson–Garnier analysis) were performed Leachman SA, Kaspar RL, Fleckman P et al. (2005) Clinical and pathological features of pachyonychia congenita. J Investig Dermatol Symp Proc using the Protein Analysis Toolkit function within the MacVector 10:3–17 9.0 software package (MacVector, Cary, NC). Liao H, Sayers JM, Wilson NJ et al. (2007a) A spectrum of mutations in keratins K6a, K16 and K17 causing pachyonychia congenita. J Dermatol CONFLICT OF INTEREST Sci 48:199–205 The authors state no conflict of interest. Liao H, Zhao Y, Baty DU et al. (2007b) A heterozygous frameshift mutation in the V1 domain of keratin 5 in a family with Dowling-Degos disease. J Invest Dermatol 127:298–300 ACKNOWLEDGMENTS Livingston RJ, Sybert VP, Smith LT et al. (2001) Expression of a truncated We thank the many patients, their families, clinical care providers, and the keratin 5 may contribute to severe palmar – plantar hyperkeratosis in referring physicians, Dr Anneli Bowen, University of Utah; Dr Roman Carlos, Guatemala; Dr Bernard Cohen, Johns Hopkins Medical Center, Baltimore; Dr epidermolysis bullosa simplex patients. J Invest Dermatol 116:970–4 Giselly De Crignis, Brazil; Dr Loretta Fiorillo, Alberta, Canada; Dr Wataru Lugassy J, Itin P, Ishida-Yamamoto A et al. (2006) Naegeli-Franceschetti- Fujimoto, Okayama, Japan; Dr Ramon Grimalt, Barcelona, Spain; Dr Anna Jadassohn syndrome and dermatopathia pigmentosa reticularis: two Hogendorf, Lo´dz, Poland; Dr Edel O’Toole, Barts and the London, Queen allelic ectodermal dysplasias caused by dominant mutations in KRT14. Mary’s School of Medicine and Dentistry, London, UK; Dr Birget Schulze, Am J Hum Genet 79:724–30 Frankfurt, Germany; Dr Michael Sorace, UCLA Department of Dermatology, McLean WH, Irvine AD, Hamill KJ et al. (2003) An unusual N-terminal LA; Dr Jose Suarez, Canary Islands, Spain; Dr Yasushi Suga, Juntendo deletion of the laminin alpha3a isoform leads to the chronic granulation University School of Medicine, Japan; Dr Jean Y. Tang, Stanford University, CA; tissue disorder laryngo-onycho-cutaneous syndrome. Hum Mol Genet Dr Maurice A.M. van Steensel, University Hospital Maastricht, The Netherlands, and Dr Mary Williams, UCSF, Department of Dermatology, San Francisco, CA, 12:2395–409 without whose participation and continued support, this study would not have McLean WHI, Hansen CD, Eliason M et al. (2011) The phenotypic and been possible. We also thank Holly Evans for her help with data preparation. molecular genetic features of pachyonychia congenita. J Invest Dermatol FJDS and NJW are supported by research grants from PC Project. (this issue)

6 Journal of Investigative Dermatology NJ Wilson et al. Pachyonychia Congenita

McLean WH, Rugg EL, Lunny DP et al. (1995) Keratin 16 and keratin 17 Smith FJD, Corden LD, Rugg EL et al. (1997) Missense mutations in mutations cause pachyonychia congenita. Nat Genet 9:273–8 keratin 17 cause either pachyonychia congenita type 2 or a Mettler G, Fraser FC (2000) Recurrence risk for sibs of children with phenotype resembling steatocystoma multiplex. J Invest Dermatol ‘‘sporadic’’ achondroplasia. Am J Med Genet 90:250–1 108:220–3 Muller FB, Anton-Lamprecht I, Kuster W et al. (1999) A premature stop codon Smith FJD, Fisher MP, Healy E et al. (2000) Novel keratin 16 mutations and mutation in the 2B helix termination peptide of keratin 5 in a German protein expression studies in pachyonychia congenita type 1 and focal epidermolysis bullosa simplex Dowling-Meara case. J Invest Dermatol palmoplantar keratoderma. Exp Dermatol 9:170–7 112:988–90 Smith FJD, Jonkman MF, van Goor H et al. (1998) A mutation in human Oh Adib C, Jones B, Liao H et al. (2008) Recurrent mutation in keratin 17 in a keratin K6b produces a phenocopy of the K17 disorder pachyonychia large family with pachyonychia congenita type 2. Arch Dermatol Res congenita type 2. Hum Mol Genet 7:1143–8 300:211–4 Sprecher E, Ishida-Yamamoto A, Becker OM et al. (2001) Evidence for novel Pfendner EG, Sadowski SG, Uitto J (2005) Epidermolysis bullosa simplex: functions of the keratin tail emerging from a mutation causing ichthyosis recurrent and de novo mutations in the KRT5 and KRT14 genes, hystrix. J Invest Dermatol 116:511–9 phenotype/genotype correlations, and implications for genetic counsel- Sprecher E, Yosipovitch G, Bergman R et al. (2003) Epidermolytic ing and prenatal diagnosis. J Invest Dermatol 125:239–43 hyperkeratosis and epidermolysis bullosa simplex caused by frameshift Richardson ES, Lee JB, Hyde PH et al. (2006) A novel mutation and large size mutations altering the v2 tail domains of keratin 1 and keratin 5. J Invest polymorphism affecting the V2 domain of keratin 1 in an African- Dermatol 120:623–6 American family with severe, diffuse palmoplantar keratoderma of the Szeverenyi I, Cassidy AJ, Chung CW et al. (2008) The Human Intermediate ichthyosis hystrix Curth-Macklin type. J Invest Dermatol 126:79–84 Filament Database: comprehensive information on a gene family Rugg EL, Rachet-Prehu MO, Rochat A et al. (1999) Donor splice site mutation involved in many human diseases. Hum Mutat 29:351–60 in keratin 5 causes in-frame removal of 22 amino acids of H1 and 1A rod Terron-Kwiatkowski A, Paller AS, Compton J et al. (2002) Two cases of domains in Dowling-Meara epidermolysis bullosa simplex. Eur J Hum primarily palmoplantar keratoderma associated with novel mutations in Genet 7:293–300 keratin 1. J Invest Dermatol 119:966–71 Smith FJ, Liao H, Cassidy AJ et al. (2005) The genetic basis of pachyonychia van Steensel MAM, Jonkman MF, van Geel M et al. (2003) Clouston syndrome congenita. J Investig Dermatol Symp Proc 10:21–30 can mimic pachyonychia congenita. J Invest Dermatol 121:1035–8 Smith FJ, Morley SM, McLean WH (2002) A novel connexin 30 mutation in Wilson NJ, Messenger AG, Leachman SA et al. (2010) Keratin K6c mutations Clouston syndrome. J Invest Dermatol 118:530–2 cause focal palmoplantar keratoderma. J Invest Dermatol 130:425–9

www.jidonline.org 7

BJD GENETICS British Journal of Dermatology The molecular genetic analysis of the expanding pachyonychia congenita case collection N.J. Wilson,1 E.A. O’Toole,2 L.M. Milstone,3 C.D. Hansen,4 A.A. Shepherd,1 E. Al-Asadi,1 M.E. Schwartz,5 W.H.I. McLean,1 E. Sprecher6 and F.J.D. Smith1 1Centre for Dermatology and Genetic Medicine, Colleges of Life Sciences and Medicine, Dentistry & Nursing, University of Dundee, Dundee, DD1 5EH, U.K. 2Blizard Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, U.K. 3Department of Dermatology, Yale University, New Haven, CT, U.S.A. 4Department of Dermatology, University of Utah, Salt Lake City, UT, U.S.A. 5PC Project, Salt Lake City, UT, U.S.A. 6Tel Aviv Sourasky Medical Center, Tel Aviv, Israel

Summary

Correspondence Background Pachyonychia congenita (PC) is a rare autosomal dominant keratinizing Frances J.D. Smith disorder characterized by severe, painful, palmoplantar keratoderma and nail dys- E-mail: [email protected] trophy, often accompanied by oral leucokeratosis, cysts and follicular keratosis. It Accepted for publication is caused by mutations in one of five keratin genes: KRT6A, KRT6B, KRT6C, KRT16 4 March 2014 or KRT17. Objectives To identify mutations in 84 new families with a clinical diagnosis of Funding sources PC, recruited by the International Pachyonychia Congenita Research Registry dur- F.J.D.S. and N.J.W. are supported by a grant ing the last few years. from the Pachyonychia Congenita Project www. Methods Genomic DNA isolated from saliva or peripheral blood leucocytes was pachyonychia.org (to F.J.D.S.). The Centre for amplified using primers specific for the PC-associated keratin genes and polymer- Dermatology and Genetic Medicine at the University of Dundee is supported by a Wellcome ase chain reaction products were directly sequenced. Trust Strategic Award (098439/Z/12/Z to Results Mutations were identified in 84 families in the PC-associated keratin genes, W.H.I.M.). comprising 46 distinct keratin mutations. Fourteen were previously unreported mutations, bringing the total number of different keratin mutations associated Conflicts of interest with PC to 105. None declared. Conclusions By identifying mutations in KRT6A, KRT6B, KRT6C, KRT16 or KRT17,this study has confirmed, at the molecular level, the clinical diagnosis of PC in these DOI 10.1111/bjd.12958 families.

What’s already known about this topic?

• Pachyonychia congenita (PC) is caused by autosomal dominant mutations in KRT6A, KRT6B, KRT6C, KRT16 or KRT17. • Plantar pain is the main symptom. • Palmoplantar keratoderma and nail dystrophy are the predominant characteristics, often accompanied by oral leucokeratosis, cysts and follicular keratosis.

What does this study add?

• This study identifies PC-associated keratin mutations in 84 new families with PC recruited by the International Pachyonychia Congenita Research Registry. • Fourteen of the 46 distinct keratin mutations were previously unreported. • This study expands the large well-phenotyped and genotyped case series of patients with PC, which is an invaluable resource for the development of mutation-specific and/or gene-specific therapies and for future clinical trials.

© 2014 The Authors. British Journal of Dermatology British Journal of Dermatology (2014) 171, pp343–355 343 published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 344 Mutations in pachyonychia congenita, N.J. Wilson et al.

Pachyonychia congenita (PC, OMIM #167200 and #167210) various types, including epidermal inclusion cysts and pilo- is a rare autosomal dominant disorder of keratinization, with sebaceous cysts; follicular keratoses; hoarseness; hyperhidrosis; hallmark signs of palmoplantar keratoderma (PPK) and nail and natal teeth. The severity and clinical features can vary, as dystrophy (Fig. 1). The main symptom is plantar pain. Addi- shown in Figure 1. While clinical case reports of PC date back tional characteristics include oral leucokeratosis; cysts of to 1906,1 and possibly earlier, the understanding behind the

K6a p.Glu472Lys K6b p.Glu472Lys K16 p.Leu132Pro

K6a p.Asn172del K16 p.Leu132Pro K16 p.Ser130del

K17 p.Met88Lys K17 p.Asn92Ser K6a p.Ile178Asn

K17 p.Asn92Ser K17 p.Met88Lys K6a p.Asn172del

Fig 1. Clinical features of pachyonychia congenita (PC). Palmoplantar keratoderma, nail dystrophy, oral leukokeratosis and cysts in patients with PC with identiﬁed mutations.

British Journal of Dermatology (2014) 171, pp343–355 © 2014 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists Mutations in pachyonychia congenita, N.J. Wilson et al. 345 genetic basis of PC was not elucidated until 1994,2 following protocol or from peripheral blood leucocytes using standard discoveries resulting in epidermolysis bullosa simplex (EBS) procedures. Samples were obtained by the IPCRR with being the first keratin disorder for which the molecular basis informed consent and ethical approval from Western Institu- – was identified.3 5 Mutations in one of five keratin genes – tional Review Board (IRB), which complies with all principles KRT6A, KRT6B, KRT6C, KRT16 or KRT17 – are now known to of the Declaration of Helsinki (Western IRB study no. – cause PC.6 9 Formerly classified as PC-1 (caused by mutations 20040468). in KRT6A or KRT16) and PC-2 (due to mutations in KRT6B or KRT17), PC nomenclature has recently been revised based on Mutation detection the molecular genetic data; for example, those with mutations in KRT6A are named PC-K6a and those with KRT16 mutations The coding regions of KRT6A, KRT6, KRT6C, KRT16 and KRT17 are PC-K16.10 were amplified using primers specific to the respective func- Keratins comprise the type I and type II groups of the inter- tional genes and to avoid amplification of K16/K17 pseudo- mediate filament subgroup of cytoskeletal proteins. The 54 genes. All primers were checked for single nucleotide polymor- known functional keratin genes are divided into epithelial and phisms using Diagnostic SNPCheck (www.ngrl.org.uk/ hair keratins.11,12 Keratin proteins, albeit very diverse in func- Manchester), and some have been modified since our previous tion, all share a similar protein structure consisting of an a-heli- publications to increase specificity (Table S1).9,15,16 For ampli- cal central rod domain comprising four domains (1A, 1B, 2A, fication of larger fragments, Takara LA Taq polymerase and 2B) connected by nonhelical regions (L1, L12, L2).13 The buffer (Takara Bio Europe/Clontech, 78100 Saint-Germain- majority of mutations causing PC are at the start and end of the en-Laye, France) was used; for smaller polymerase chain reac- a-helical rod domain (within 1A and 2B) within the helix tion (PCR) reactions, HotStarTaq DNA Polymerase and buffer boundary motif domains that show evolutionary conservation system (Qiagen, Crawley, U.K.) was used according to the man- throughout all intermediate filaments. These motif regions are ufacturer’s instructions. Specific PCR conditions for each primer thought to play a vital role in the interaction of filaments during set are available on request. PCR products were purified using cytoskeleton construction in forming end–end overlap inter- QiaQuick PCR spin columns (Qiagen) or ExoSAP (using exonu- actions during filament assembly.14 clease 1 and shrimp antartic phosphatase) and sequenced using More than 20 keratin genes are now known to cause internal primers on an ABI 3700 Automated DNA sequencing directly or predispose to fragility syndromes (www.interfil. machine (Applied Biosystems, Foster City, CA, U.S.A.). Previ- org), emphasizing the importance of keratins in skin and epi- ously unreported mutations were excluded from at least 90 con- thelial appendages. Keratin diseases are almost always inherited trol DNA samples (180 chromosomes) by sequencing or in an autosomal dominant manner where mutations act in a restriction enzyme digests. The occurrence of sequence changes dominant negative way, disrupting the ability of keratin fila- was also checked on the Exome Variant Server (http://evs.gs. ments to form a strong cytoskeleton, leaving the skin in a washington.edu/EVS) and dbSNP database. fragile, unprotected condition unable to withstand trauma. The mutational spectrum covers all types (missense mutations, RNA extraction small deletion/insertion mutations, splice-site mutations and nonsense mutations), and the disease produced is confined A 3-mm punch biopsy was taken from an affected individual strictly to the epithelial site where the mutant keratin protein from family 17 with splice-site mutation K6a c.541–2A>G is normally expressed; for example, keratin K6a is expressed and put in RNAlater (Life Technologies, Paisley, U.K.). mRNA in palmoplantar epidermis, nail epithelia and mucosal tissues was extracted using a Quick Prep Micro mRNA purification kit – the tissues affected in PC. (GE Healthcare UK, Little Chalfont, U.K.) and reverse tran- The International Pachyonychia Congenita Research Registry scribed using AMV Reverse Transcriptase (Promega, South- (IPCRR) has, to date, collected clinical and molecular data from ampton, U.K.). cDNA was amplified using the Expand High more than 500 patients with PC and their families (~270 fami- Fidelity PCR system (Roche, Mannheim, Germany) with prim- lies) worldwide. Here, we report molecular analysis of 84 fami- ers specific for K6a within exon 1 through exon 5. lies registered within the last few years of this expanding case collection with a clinical diagnosis of PC. Mutations were identi- Cloning of polymerase chain reaction fragments fied in KRT6A, KRT6B, KRT6C, KRT16 or KRT17. Forty-six distinct keratin mutations were found, 14 of which were previously The splice-site mutation K6a c.541–2A>G was confirmed by unreported mutations, bringing the total number of different cloning the PCR product derived from cDNA (above) into keratin mutations associated with PC to 105. pCR2.1 vector (TA cloning kit; Life Technologies). Several independent clones were sequenced. Materials and methods K16 constructs and transfections Genomic DNA was extracted from saliva collected in an Oragene DNA sample collection kit (DNA Genotek, Kanata, Full-length K16 cDNA was cloned into expression vector ON, Canada) and extracted according to the manufacturer’s pCR3.1 (Life Technologies) and used as a template for site-

© 2014 The Authors. British Journal of Dermatology British Journal of Dermatology (2014) 171, pp343–355 published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists 346 Mutations in pachyonychia congenita, N.J. Wilson et al. directed mutagenesis (Stratagene, Cedar Creek, TX, U.S.A.) to c.1417dupG. This results in a frameshift starting with codon make constructs containing K16 p.Arg418Cys and K16 473, which is changed from glutamic acid to a glycine residue p.Arg418Pro. The three plasmids were transfected individually and creates a premature stop codon at position 91 of the new into PtK2 cells using Lipofectamine 2000 (Life Technologies), reading frame; this is predicted to cause loss of normal protein according to the manufacturer’s protocol. Cells were fixed at 48 function through protein truncation. Specifically, the last 92 cor- and 72 h post-transfection in 1 : 1 methanol : acetone for rect amino acids are replaced by 90 incorrect amino acids, which 5 min, air dried and double-label immunofluorescence was per- are very different in sequence leading to a foreign protein formed. Transfected K16 was detected with rabbit polyclonal (Fig. 2). A previously reported mutation, K6a p.Ser505Glnf- antisera (1 : 500 dilution) against human K16 (gift of Pierre sTer59, downstream of the above mutation, also resulted in a 1- Coulombe, Johns Hopkins University, Baltimore, MD, U.S.A.) bp insertion,16 this time in the V2 domain, and the resulting 58 and the endogenous K8 with neat supernatant monoclonal anti- foreign amino acids were the same as in the above mutation. body LE41 to PtK2 K8 (gift of Birgitte Lane, Institute of Medical A previously unreported nonsense mutation, K6a p.Glu461- Biology, Immunos, Singapore). Alexafluor 488 goat anti-mouse Ter (in Family 21) at the start of the helix termination motif and Alexafluor 594 goat anti-rabbit secondary antibodies (Life in the 2B domain of K6a, results in a premature stop codon Technologies) were used at a dilution of 1 : 1000. Nuclei were leading to a truncated protein lacking the helix termination stained with 40,6-diamidino-2-phenylindole (DAPI). motif and tail domain. Missense mutations have previously been reported at this position: K6a p.Glu461, p.Glu461Lys Results and p.Glu461Gln (www.interfil.org). Splice-site mutations were found in six families in KRT6A. In five of these (families 16–20) mutations were at the intron Clinical details 1/exon 2 boundary of K6a; three different sequence changes The main clinical features of individuals involved in this study were identified, all of which we have reported previously,16 are reported in Table 1. All cases were recruited through the but at that time, owing to the unavailability of mRNA, we IPCRR, an ongoing research programme to identify patients were unable to identify the mutations at the protein level. with PC worldwide and to collect detailed clinical and molecu- However, in this study we obtained mRNA from a skin biopsy lar data from all individuals registered. Autosomal dominant from an affected individual from family 17 with mutation K6a inheritance was observed in 49 of 84 families, while the c.541–2A>G. The resulting cDNA was amplified by PCR and remaining cases were apparently due to spontaneous muta- cloned, and several clones were sequenced to identify the con- tions. In many cases with previously reported mutations the sequence of this mutation. The mutation is predicted to result parents were not screened. in an in-frame deletion of the first six amino acids of exon 2, K6a p. Val181_Gln186del. mRNA was unavailable from other families with the different sequence changes at this splice site. Mutation analysis In a sixth family (family 31) a previously unreported splice- Of the 84 families, we identified previously unreported het- site mutation was identified at the exon 8/intron 9 boundary erozygous mutations in 14 of them and known heterozygous (c.1460–1G>C); unfortunately, we were unable to obtain mutations in 70 families. The majority were missense muta- mRNA from this family to determine the effect of this geno- tions, with the remainder being small in-frame deletion, mic mutation on RNA splicing. frameshift, nonsense or splice-site mutations in KRT6A, KRT6B, KRT6C, KRT16 or KRT17 (Table 1). Mutations in KRT6B (PC-K6b)

Mutations in KRT6B were found in nine families, of which one Mutations in KRT6A (PC-K6a) was novel. Only three different mutations have previously Thirty-one families had mutations in KRT6A and of the 18 dif- been reported in KRT6B17 (www.interfil.org): p.Asn172del, ferent mutations, four, including a nonsense mutation, were p.Glu461Lys and p.Glu472Lys. Here, eight of the nine families previously unreported. The most commonly reported mutation with mutations in KRT6B had mutations at these sites, further in PC is K6a p.Asn172del, found in 10 families here, and indicating that these are mutation hotspot residues in KRT6B: overall in ~30% of kindreds with mutations in K6a and in p.Asn172del (families 32–35), p.Glu461Lys (family 36) and ~13% of all families with a PC mutation (including published, p.Glu472Lys (families 38–40). Family 37 was found to have www.interfil.org, and those reported here). an unreported missense mutation within the 2B domain of A novel missense mutation, K6a p.Phe174Ile, was identified KRT6B, p.Leu469Arg. The analogous mutation has been in family 13; other amino acid substitutions at this residue reported in KRT6A (www.interfil.org). have been reported, p.Phe174Ser in at least 12 families, including one in this study, and p.Phe174Cys in one family.17 Mutations in KRT6C (PC-K6c) Mutation K6a p.Glu473GlyfsTer91 (family 30), an unreported frameshift mutation at the end of the 2B domain of Only two families (families 41 and 42) had mutations in K6a, is due to duplication of a single ‘G’ nucleotide, KRT6C, both with the same known mutation, p.Glu472Lys.

04TeAuthors. The 2014 Table 1 Mutations and clinical details in new cases of pachyonychia congenita

rts ora fDermatology of Journal British 1 K6a p.Glu163Lys c.487G A Known Familial Somewhat painful PPK All 20 Yes NA No 2 K6a p.Leu170Phe c.508C>T Known Spontaneous Not painful PPK, very mild All 20 No NA No 3 K6a p.Asn172del c.516_518delCAA Known Familial Very painful, but does PK All 20 No NA No not use medication 4 K6a p.Asn172del c.516_518delCAA Known Familial Very painful, but does PPK All 20 Yes NA No not use medication 5 K6a p.Asn172del c.516_518delCAA Known Spontaneous NA; under 2 years old NA; under 10 ﬁngernails, Yes NA No 2 years old 2 toenails 6 K6a p.Asn172del c.516_518delCAA Known Spontaneous Very painful, but does PPK All 20 Yes Pilosebaceous, No not use medication follicular hyperkeratosis 7 K6a p.Asn172del c.516_518delCAA Known Spontaneous Very painful, but does PPK All 20 Yes Steatocystoma, No not use medication pilosebaceous, follicular hyperkeratosis 8 K6a p.Asn172del c.516_518delCAA Known Spontaneous Somewhat painful PPK All 20 Yes Steatocystoma, No follicular hyperkeratosis 9 K6a p.Asn172del c.516_518delCAA Known Familial Not painful PPK All 20 Yes Follicular No hyperkeratosis Wilson N.J. congenita, pachyonychia in Mutations 10 K6a p.Asn172del c.516_518delCAA Known Spontaneous Often requires medication PPK All 20 Yes NA No

rts ora fDraooy(2014) Dermatology of Journal British to handle the pain 11 K6a p.Asn172del c.516_518delCAA Known Spontaneous Somewhat painful PPK All 20 Yes NA No 12 K6a p.Asn172del c.516_518delCAA Known Familial Often requires medication PPK All 20 Yes Follicular No to handle the pain hyperkeratosis 13 K6a p.Phe174Ile c.520T>A Unreported Spontaneous Very painful, but does not PK All 20 Yes Pilosebaceous, No use medication follicular hyperkeratosis 14 K6a p.Phe174Ser c.521T>C Known Spontaneous Very painful, but does PK All 20 Yes NA No not use medication 15 K6a p.Ile178Asn c.533T>A Known Spontaneous Very painful, but does PK All 20 Yes Pilosebaceous, No not use medication follicular hyperkeratosis 16 ND c.541–1G>C Known Familial Often requires medication PK 6 ﬁngernails, Yes NA No 171 to handle the pain 10 toenails pp343–355 , – >

17 K6a p.Val181_Gln186del c.541 2A G Known Familial Somewhat painful PK All 20 Yes Pilosebaceous, No al. et follicular hyperkeratosis 347 348 rts ora fDraooy(2014) Dermatology of Journal British Table 1 (continued) uain npcynci ognt,NJ Wilson N.J. congenita, pachyonychia in Mutations Cysts and/or Born Mutation – protein Unreported Familial or Palmoplantar Thickened Oral follicular with Family change DNA change or known spontaneous Plantar pain keratoderma nails leucokeratosis hyperkeratosis teeth? 18 K6a p.Val181_Gln186del c.541–2A>G Known Familial Very painful, but does PPK All 20 Yes Pilosebaceous, No not use medication follicular hyperkeratosis 19 K6a p.Val181_Gln186del c.541–2A>G Known Familial Very painful, but does PK All 20 Yes NA No not use medication 20 ND c.541–2A>C Known Spontaneous Somewhat painful PK 5 ﬁngernails, Yes NA No 10 toenails 171 21 K6a p.Glu461Ter c.1381G>T Unreported Spontaneous Very painful, but does PK 5 ﬁngernails, Yes Steatocystoma, No pp343–355 , not use medication 10 toenails pilosebaceous, follicular hyperkeratosis 22 K6a p.Ile462Asn c.1385T>A Known Spontaneous Very painful, but does PK All 20 Yes Pilosebaceous, No not use medication follicular tal. et ulse yJh ie osLdo eafo rts soito fDermatologists of Association British of behalf on Ltd Sons & Wiley John by published hyperkeratosis 23 K6a p.Ala463Pro c.1387G>C Known Spontaneous Somewhat painful PPK All 20 Yes Follicular No hyperkeratosis 24 K6a p.Thr464Pro c.1390A>C Known Familial Very painful, but does PK All 20 Yes Follicular No not use medication hyperkeratosis 25 K6a p.Leu468Pro c.1403T>C Known Spontaneous Very painful, but does PPK All 20 Yes Follicular No not use medication hyperkeratosis 26 K6a p.Leu469Pro c.1406T>C Known Familial Often requires medication PPK All 20 Yes Steatocystoma, No to handle the pain pilosebaceous 27 K6a p.Leu469Pro c.1406T>C Known Spontaneous Often requires medication PPK All 20 Yes Follicular No to handle the pain hyperkeratosis 28 K6a p.Glu472Lys c.1414G>A Known Spontaneous Not painful (under 3 years) PK All 20 No Small red bumps No

04TeAuthors. The 2014 spots on his face, but they could be standard baby acne 29 K6a p.Glu472Lys c.1414G>A Known Spontaneous Very painful, but does PPK All 20 Yes Steatocystoma, No not use medication pilosebaceous, follicular

rts ora fDermatology of Journal British hyperkeratosis 30 K6a p.Glu473GlyfsTer91 c.1417dupG Unreported Spontaneous Somewhat painful PPK 10 fingernails, Yes NA No 7 toenails 31 ND c.1460–1G>C Unreported Familial Often requires medication PPK 2 fingernails, No NA No to handle the pain 10 toenails 32 K6b p.Asn172del c.516_518delCAA Known Familial Somewhat painful PK 10 fingernails, Yes NA No 7 toenails ulse yJh ie osLdo eafo rts soito fDermatologists of Association British of behalf on Ltd Sons & Wiley John by published © Table 1 (continued) 04TeAuthors. The 2014

Cysts and/or Born Mutation – protein Unreported Familial or Palmoplantar Thickened Oral follicular with Family change DNA change or known spontaneous Plantar pain keratoderma nails leucokeratosis hyperkeratosis teeth? 33 K6b p.Asn172del c.516_518delCAA Known Familial Very painful, but does PPK 6 ﬁngernails, Yes Steatocystoma, No

rts ora fDermatology of Journal British not use medication 10 toenails pilosebaceous 34 K6b p.Asn172del c.516_518delCAA Known Familial Very painful, but does PK 6 ﬁngernails, No Pilosebaceous, No not use medication 10 toenails follicular hyperkeratosis 35 K6b p.Asn172del c.516_518delCAA Known Familial Somewhat painful PPK 8 ﬁngernails, No Pilosebaceous, No 10 toenails follicular hyperkeratosis 36 K6bGlu461Lys c.1381G>A Known Spontaneous Very painful, but does PPK 7 toenails Yes Follicular No not use medication hyperkeratosis 37 K6b p.Leu469Arg c.1406T>G Unreported Spontaneous Somewhat painful PK 6 toenails Yes Follicular No hyperkeratosis 38 K6b p.Glu472Lys c.1414G>A Known Spontaneous Very painful, but does PK 8 toenails Yes NA No not use medication 39 K6b p.Glu472Lys c.1414G>A Known Familial Somewhat painful PK 4 toenails No Steatocystoma, No pilosebaceous, follicular hyperkeratosis 40 K6b p.Glu472Lys c.1414G>A Known Spontaneous Very painful, but does PPK 9 toenails No Steatocystoma No not use medication 41 K6c p.Glu472Lys c.1414G>A Known Familial Very painful, but does PPK 4 toenails Yes Steatocystoma, No Wilson N.J. congenita, pachyonychia in Mutations not use medication pilosebaceous >

rts ora fDraooy(2014) Dermatology of Journal British 42 K6c p.Glu472Lys c.1414G A Known Familial Somewhat painful PK 2 toenails No NA No 43 K16 p.Met121Lys c.362T>A Known Familial Very painful, but does PPK All 20 No NA No not use medication 44 K16 p.Met121Thr c.362T>C Known Spontaneous Very painful, but does PK 4 toenails No NA No not use medication 45 K16 p.Leu124His c.371T>A Known Familial Very painful, but does PPK 6 ﬁngernails, No NA No not use medication 10 toenails 46 K16 p.Leu124His c.371T>A Known Familial Very painful, but does PPK 5 toenails No NA No not use medication 47 K16 p.Asn125Asp c.373A>G Known Familial Very painful, but does PPK 10 toenails No Pilosebaceous No not use medication 48 K16 p.Asn125Ser c.374A>G Known Spontaneous Very painful, but does PPK All 20 No NA No not use medication 171 49 K16 p.Asn125Ser c.374A>G Known Spontaneous Somewhat painful; only PK 2 toenails No NA No pp343–355 ,

6 years old al. et 50 K16 p.Asn125Ser c.374A>G Known Familial Very painful, but does PPK 3 toenails Yes NA No not use medication 349 350 rts ora fDraooy(2014) Dermatology of Journal British uain npcynci ognt,NJ Wilson N.J. congenita, pachyonychia in Mutations Table 1 (continued)

Cysts and/or Born Mutation – protein Unreported Familial or Palmoplantar Thickened Oral follicular with Family change DNA change or known spontaneous Plantar pain keratoderma nails leucokeratosis hyperkeratosis teeth? 51 K16 p.Asn125Ser c.374A>G Known Familial Somewhat painful PPK 8 toenails No NA No 52 K16 p.Asn125Ser c.374A>G Known Spontaneous Somewhat painful PPK 2 toenails No Follicular No hyperkeratosis 53 K16 p.Asn125Ser c.374A>G Known Familial Often requires medication PPK All 20 Yes Follicular No to handle the pain hyperkeratosis 171 54 K16 p.Arg127Pro c.380G>C Known Familial Somewhat painful PPK All 20 No Follicular No pp343–355 , hyperkeratosis 55 K16 p.Arg127Gly c.379C>G Unreported Familial Very painful, but does PPK All 20 No Steatocystoma No not use medication 56 K16 p.Arg127Cys c.379C>T Known Familial Often requires medication PK 5 toenails No Follicular No to handle the pain hyperkeratosis tal. et ulse yJh ie osLdo eafo rts soito fDermatologists of Association British of behalf on Ltd Sons & Wiley John by published 57 K16 p.Arg127Cys c.379C>T Known Familial Very painful, but does PPK 2 toenails Yes NA No not use medication 58 K16 p.Arg127Cys c.379C>T Known Familial Very painful, but does PPK 3 toenails No NA No not use medication 59 K16 p.Arg127Cys c.379C>T Known Familial Very painful, but does PPK None No NA No not use medication 60 K16 p.Arg127Cys c.379C>T Known Familial Very painful, but does PK 6 toenails No NA No not use medication 61 K16 p.Arg127Ser c.379C>A Unreported Familial Often requires medication PK 7 toenails No NA No to handle the pain 62 K16 p.Arg127His c.381G>A Unreported Familial Very painful, but does PPK 4 toenails Yes Pilosebaceous, No not use medication follicular

04TeAuthors. The 2014 63 K16 p.Ser130del c.389–391delCCT Known Familial Often requires medication PPK All 20 No Follicular No to handle the pain hyperkeratosis 64 K16 p.Leu132Pro c.395T>C Known Familial Very painful, but does PPK All 20 No NA No not use medication 65 K16 p.Arg418Pro c.1253G>C Unreported Familial Very painful, but does PK 2 toenails No NA No not use medication

rts ora fDermatology of Journal British 66 K16 p.Arg418_Arg419del c.1253_1258delGCCGCC Unreported Familial Very painful, but does PPK 2 ﬁngernails, Yes NA No not use medication 3 toenails 67 K17 p.Met88Lys c.263T>A Known Spontaneous NA None (under All 20 No Follicular Yes 3 years of age) hyperkeratosis ulse yJh ie osLdo eafo rts soito fDermatologists of Association British of behalf on Ltd Sons & Wiley John by published ©

04TeAuthors. The 2014 Table 1 (continued)

Cysts and/or Born Mutation – protein Unreported Familial or Palmoplantar Thickened Oral follicular with Family change DNA change or known spontaneous Plantar pain keratoderma nails leucokeratosis hyperkeratosis teeth? 68 K17 p.Met88Arg c.263T>G Unreported Spontaneous Very painful, but does PPK 8 ﬁngernails, Yes Steatocystoma, No

rts ora fDermatology of Journal British not use medication 10 toenails pilosebaceous, follicular hyperkeratosis 69 K17 p.Asn90_Asp93delinsIle c.269_278del10insT Unreported Familial Very painful, but does PPK All 20 No Follicular No not use medication hyperkeratosis 70 K17 p.Leu91Pro c.272T>C Unreported Spontaneous Very painful, but does PPK All 20 No Steatocystoma, No not use medication follicular hyperkeratosis 71 K17 p.Asn92Ser c.275A>G Known Familial Very painful, but does PK All 20 No Steatocystoma, Yes not use medication pilosebaceous, follicular hyperkeratosis 72 K17 p.Asn92Ser c.275A>G Known Familial Not painful PPK 6 fingernails, No Steatocystoma, Yes 10 toenails pilosebaceous, follicular hyperkeratosis 73 K17 p.Asn92Ser c.275A>G Known Spontaneous Not painful PPK 6 fingernails, No Steatocystoma, Yes 10 toenails pilosebaceous 74 K17 p.Asn92Ser c.275A>G Known Spontaneous Often requires medication PK 10 fingernails, Yes Pilosebaceous No uain npcynci ognt,NJ Wilson N.J. congenita, pachyonychia in Mutations to handle the pain 8 toenails 75 K17 p.Asn92Ser c.275A>G Known Spontaneous NA None (under All 20 No NA Yes

rts ora fDraooy(2014) Dermatology of Journal British 3 years of age) 76 K17 p.Asn92Ser c.275A>G Known Familial Somewhat painful PPK 6 ﬁngernails, No Steatocystoma, Yes 2 toenails pilosebaceous, follicular hyperkeratosis 77 K17 p.Asn92Ser c.275A>G Known Familial Often requires medication PPK 2 ﬁngernails, No Pilosebaceous, Yes to handle the pain 10 toenails follicular hyperkeratosis 78 K17 p.Arg94Ser c.280C>A Unreported Familial Very painful, but does PK 3 toenails No Steatocystoma, No not use medication pilosebaceous, follicular hyperkeratosis

171 79 K17 p.Arg94Cys c.280C>T Known Familial Very painful, but does PPK 4 toenails No Steatocystoma, No

pp343–355 , not use medication pilosebaceous,

follicular al. et hyperkeratosis 351 352 Mutations in pachyonychia congenita, N.J. Wilson et al.

Mutations in KRT16 (PC-K16)

Mutations in KRT16 were found in 24 families; the most com- Yes No Born with teeth? Yes mon recurrent mutations identified were K16 p.Asn125Ser, p.Arg127Cys and p.Leu132Pro (www.interfil.org). Five of the 14 distinct K16 mutations were previously unreported. The 20 or more different mutations reported in keratin 16 include p.Arg127Cys and p.Arg127Pro. In this cohort, we identified pilosebaceous, follicular hyperkeratosis hyperkeratosis previously unreported mutations in three families (families Steatocystoma, Steatocystoma Yes Cysts and/or follicular hyperkeratosis NA Pilosebaceous Yes 55, 61 and 62) at K16 p.Arg127: p.Arg127Gly, p.Arg127Ser and p.Arg127His. We also identified one additional family (family 54) with mutation p.Arg127Pro and five families (families 56–60) with p.Arg127Cys. Interestingly, the pheno- Oral leucokeratosis typic variation seen with these latter two mutations indicates a genotype–phenotype correlation, as proposed in a small known with certainty). study.18 Fu et al. showed that individuals with mutation p.Arg127Cys were more likely to have milder features of the All 20 No 1 toenail No Thickened nails All 20All 20 No No All 20 Yes Follicular disorder (milder nail changes, less palmoplantar pain, etc.) than those with amino acid substitution p.Arg127Pro.18 Including these additional mutations, ~24% of reported mutations in KRT16 occur at this p.Arg127 (www.interfil.org). Interestingly, a proline substitution for arginine was found 2 years old PPK PK NA Palmoplantar keratoderma to be deleterious in another portion of the helix, whereas a cysteine substitution was not. In Family 65, we identified a new mutation K16 p.Arg418Pro. This mutation in the helix termination motif domain of K16 is predicted to be pathogenic. The unaffected parents of the proband were both wild type. Another sequence change at this position, p.Arg418Cys, which might be predicted to be pathogenic owing to its position within the helix termination motif, is listed on the dbSNP not use medication not use medication database (minor allele frequency from 1000 Genomes project = 19%) and also on the Exome Variant Server (http://evs.gs. washington.edu/EVS/), indicating that it may be a nonpatho- genic sequence change. We also found p.Arg418Cys in one of Familial or spontaneous Plantar pain 92 anonymous unrelated control DNA samples and in some members of three unrelated PC families in which known mutations were identified in KRT6A, KRT16 or KRT17. To inves- Known Spontaneous Very painful, but does Known SpontaneousKnown NA; under 2 years old Familial NA; Very under painful, but does Known Familial Somewhat painful PPK Known Familial Not painful Unreported or known tigate the effect of these sequence changes at K16 p.Arg418, PtK2 cells were transfected with plasmids expressing either wild-type K16, K16 p.Arg418Cys or K16 p.Arg418Pro cDNAs. At 48 h and 72 h post-transfection cells were fixed, and stained with a polyclonal antibody against K16 to detect transfected K16. Endogenous K8 was detected with monoclonal > T > C > C > C > C antibody LE41 to PtK2 K8. Nuclei were stained with DAPI. Cells were examined for those containing filaments, filaments c.296T c.284T c.284T c.296T plus aggregates or only aggregates. In cells transfected with wild-type K16, 895% of transfected cells showed a defined keratin cytoskeleton where wild-type K16 co-localized with endogenous K8; the remaining 105% of transfected cells had filaments plus aggregates. Cells transfected with K16 p.Arg418Cys showed a similar pattern to wild-type K16, with 89% of transfected cells with normal filament network and 11% of cells with filaments plus aggregates. However, in cells (continued) Mutation – protein change DNA change transfected with K16 p.Arg418Pro, 985% of cells contained aggregates and there was collapse/aggregation of the endoge- PK, plantar keratoderma; PPK, palmoplantar keratoderma; NA, not applicable; ND, not determined (protein consequences of a splice site mutation not 84 K17 p.Leu99Pro 8081 K17 p.Arg94Cys K1782 p.Leu95Pro K17 p.Leu95Pro c.280C 83 K17 p.Leu99Pro Table 1 Family nous network; only 15% of transfected cells showed a normal

(a)

(b)

Fig 2. Kyle–Doolittle hydrophilicity analysis of normal and mutant K6a. (a) Normal K6a tail domain consists of alternating hydrophobic and hydrophilic sequences compared with (b) the mutant tail domain, K6a p.Glu473GlyfsTer91, which is mainly hydrophilic.

filament network (Fig. S1). These results demonstrate that the been previously found at this residue p.M88, p.Met88Lys and mutation p.Arg418Pro is disruptive to the normal filament p.Met88Thr) and p.Leu91Pro (family 70). A novel insertion/ network formation, which, in vivo, would result in PC, whereas deletion mutation in K17, p.Asn90_Asp93delinsIle, found in p.Arg418Cys, in vitro, produces a normal filament network Family 69, results in deletion of 10 nucleotides and insertion similar to wild-type K16 and is therefore unlikely to be patho- of one nucleotide (T). This leads to an in-frame deletion in a genic. critical region within the helix initiation motif. Another novel mutation identified in K16 is a 6-bp in- frame deletion mutation, K16 p.Arg418_Arg419del, in family Discussion 66, which results in the deletion of two amino acids from the helix termination motif. The mutations identified in these 84 families add to the increasing data set of the IPCRR of detailed clinical information and corresponding molecular data of individuals with PC. In all Mutations in KRT17 (PC-K17) cases, the mutations were heterozygous sequence changes – KRT17 mutations were found in 18 families and included four missense, nonsense, small deletion/insertion or splice-site novel mutations. The most common known mutation found mutations – confirming this is predominantly, if not exclusively, was K17 p.Asn92Ser; including previous publications (www. an autosomal dominant disorder. While there are a few case interfil.org) and this study, it occurs in ~8% of PC families reports of recessive PC reported in the literature,19 as yet there with a confirmed mutation. are no recessive cases with confirmed genetic testing. Three previously unreported missense mutations were iden- The majority of the mutations identified have been previously tified in KRT17. Family 78 was found to harbour a mutation reported with some mutations, for example K6a p.Asn172del, in K17, p.Arg94Ser. p.Arg94 is a residue already known to be K16 p.Asn125Ser, K16 p.Arg127Cys and K17 p.Asn92Ser occur- susceptible to mutation owing to several reported cases result- ring frequently. For development of mutation-specific forms of ing in amino acid substitutions p.Arg94Cys, p.Arg94His and treatment, for example small interfering RNA, these residues p.Arg94Pro (www.interfil.org). Two other unreported muta- that are commonly mutated would be obvious targets owing to tions in K17 were p.Met88Arg (family 68) (mutations have the larger number of patients that could potentially be treated.

More than a third of cases analysed in this study had mutations mutations in connexin, 30 present with nail dystrophy, some in K6a, which may partly be reflected by the fact that individuals PPK and alopecia. The severity of the alopecia varies widely with K6a mutations tend to present with the most severe fea- from very subtle to total alopecia but is a key indicator of tures of PC and are therefore more likely to search/ask for sup- Clouston syndrome due to mutations in connexin 30, rather port, find the PC Project website (www.pachyonychia.org) and than a diagnosis of PC. Autosomal dominant mutations in join the IPCRR. Conversely, only two families were identified DSG121 and DSP, typically result in striate PPK,25 although this with mutations in KRT6C. In family 42, the proband had for can also occur in some patients with PC. many years been diagnosed with EBS due to blistering of palms Autosomal recessive cases due to desmoplakin mutations and soles and subsequent hyperkeratosis. From the four cases also have features that overlap with PC, including palmoplan- previously reported with KRT6C mutations,17 together with tar blistering and keratoderma with nail dystrophy, which can those in this report, it appears that these individuals present lead to misdiagnosis. However, these individuals also have dis- with a milder clinical phenotype than those with mutations in tinctive sparse, woolly hair and, importantly, they are at risk KRT6A, KRT6B, KRT16 or KRT17. Information regarding the phe- of cardiomyopathy. Therefore, diagnosis at the molecular level notype of those with mutations in KRT6C is limited due to the is important for these individuals in defining the risk of car- small number of reported cases but this could be explained as diomyopathy and to allow appropriate monitoring of their only rarely do they come to clinical attention. condition. The results reported here, together with previously pub- Mutations in FZD6 were recently discovered as the cause of lished data, are summarized in Figure 3, which shows the autosomal recessive nail dysplasia.23 These individuals present spectrum of mutations in the five keratin genes associated with 20-nail dystrophy from birth or shortly afterwards in the with PC: KRT6A, KRT6B, KRT6C, KRT16 and KRT17. form of thickened, discoloured, claw-shaped nails. There is no The overlapping phenotype of inherited PPKs, independent involvement of other ectodermal tissues. FZD6 screening of their genotype, can make diagnosing them clinically a con- should be considered for any spontaneous cases or known fusing field. recessive cases with isolated nail dystrophy. Individuals presenting with symptoms of PC should Genetic analysis of the cases in this study has confirmed undergo genetic analysis for mutations in the five known PC their clinical diagnosis of PC. A small number of cases within genes (KRT6A, KRT6B and KRT6C, KRT16 and KRT17). In addi- the IPCRR (not reported here) that are clinically not typical of tion, other candidate genes may need to be considered, PC have been shown to have mutations in the other genes including GJB6 encoding connexin 30,20 desmoglein 1 mentioned above. Correct molecular diagnosis is important to (DSG1),21 desmoplakin (DSP),22 keratin 9 (KRT9),17 and friz- aid in appropriate genetic counselling and patient care. This zled 6,23,24 mutations in that mimic some features of the PC large, well-phenotyped and genotyped case series is an invalu- phenotype and hence should be encompassed into the able resource for the development of mutation-specific and/or differential diagnosis of PC. For example, individuals with gene-specific therapies, and for future clinical trials.

Acknowledgments

We thank all the patients and families involved in this study. We also thank all the referring physicians: Dr Susan J. Bayliss, Washington University School of Medicine, St Louis, MO, U.S.A.; Dr Jeffrey F. Corpuz, University of Santo Tomas Hospi- KRT17 KRT6A tal, Manila, Philippines; Dr Inacio Rezende Faver, IR Faver 24% 44% Clinic, Miracema-RJ, Brazil; Dr Celia Posada Garcia, Hospital Provincial Dermatology Service, Pontevedra, Spain; Dr Ros- amaria Gomes, Macei, Alagoas, Brazil; Dr Alan Irvine, Our Lady’s Hospital for Sick Children, Dublin, Ireland; Dr Vanesa KRT16 Lopez Gonzalez, Unidad de Genetica Medica, Murcia, Spain; 25% Dr K. Kempcke, Universit€ats-Hautklinik Kiel, Kiel, Germany; Dr Rudolph Leube, Institute of Molecular and Cellular Anat- omy, Germany; Dr Helen Lewis, Queen Elizabeth Hospital, KRT6B 5% Birmingham, U.K.; Dr Jemima E. Mellerio, Guy’s & St. Tho- mas’ Hospital, London, U.K.; Dr Elena Montalvan Miro, Uni- KRT6C 2% versity of Puerto Rico School of Medicine, San Juan, Puerto Rico; Dr Megan Mowbray, Queen Margaret Hospital, Dun- Fig 3. Spectrum of mutations causing pachyonychia congenital (PC), fermline, U.K.; Dr Anders Vahlquist, Uppsala University Hos- showing the percentage of families in this study and previous pital, Sweden; Dr Theresa A. Scholz, Accent Dermatology and publications with mutations in the ﬁve keratin genes associated with Laser Institute, Lakewood, CO, U.S.A.; Dr Yasushi Suga, PC: KRT6A, KRT6B, KRT6C, KRT16 and KRT17. Juntendo University, Chiba, Japan; and Dr Wujanto, Luton

Hospital, U.K. Thanks also to Holly Evans of PC Project for all 18 Fu T, Leachman SA, Wilson NJ et al. Genotype–phenotype correla- her help with data preparation. tions among pachyonychia congenita patients with K16 mutations. J Invest Dermatol 2011; 131:1025–8. 19 Haber RM, Rose TH. Autosomal recessive pachyonychia congenita. References Arch Dermatol 1986; 122:919–23. 20 van Steensel MAM, Jonkman MF, van Geel M et al. Clouston syn- 1 Jadassohn J, Lewandowski F. [Pachyonychia Congenita: Keratosis Dissemi- drome can mimic pachyonychia congenita. J Invest Dermatol 2003; nata Circumscripta (Follicularis). Tylomata. Leukokeratosis Linguae.] Berlin: 121:1035–8. Urban and Schwarzenberg, 1906. 21 Rickman L, Imrak D, Stevens HP et al. N-terminal deletion in a 2 Munro CS, Carter S, Bryce S et al. A gene for pachyonychia con- desmosomal cadherin causes the autosomal dominant skin disease genita is closely linked to the keratin gene cluster on 17q12–q21. striate palmoplantar keratoderma. Hum Mol Genet 1999; 8:971–6. J Med Genet 1994; 31:675–8. 22 Smith FJ, Wilson NJ, Moss C et al. Compound heterozygous muta- 3 Lane EB, Rugg EL, Navsaria H et al. A mutation in the conserved tions in desmoplakin cause skin fragility and woolly hair. Br J helix termination peptide of keratin 5 in hereditary skin blistering. Dermatol 2012; 166:894–6. Nature 1992; 356:244–6. 23 Frojmark AS, Schuster J, Sobol M et al. Mutations in Frizzled 6 4 Vassar R, Coulombe PA, Degenstein L et al. Mutant keratin expres- cause isolated autosomal-recessive nail dysplasia. Am J Hum Genet sion in transgenic mice causes marked abnormalities resembling a 2011; 88:852–60. human genetic skin disease. Cell 1991; 64:365–80. 24 Wilson NJ, Hansen CD, Azkur D et al. Recessive mutations in the 5 Bonifas JM, Rothman AL, Epstein EH. Epidermolysis bullosa sim- gene encoding frizzled 6 cause twenty nail dystrophy – expanding plex: evidence in two families for keratin gene abnormalities. the differential diagnosis for pachyonychia congenita. J Dermatol Sci Science 1991; 254:1202–5. 2013; 70:58–60. 6 Bowden PE, Haley JL, Kansky A et al. Mutation of a type II keratin 25 Armstrong DK, McKenna KE, Purkis PE et al. Haploinsufficiency of gene (K6a) in pachyonychia congenita. Nat Genet 1995; 10:363–5. desmoplakin causes a striate subtype of palmoplantar keratoderma. 7 McLean WHI, Rugg EL, Lunny DP et al. Keratin 16 and keratin 17 Hum Mol Genet 1999; 8:143–8. mutations cause pachyonychia congenita. Nat Genet 1995; 9:273–8. 8 Smith FJD, Jonkman MF, van Goor H et al. A mutation in human keratin K6b produces a phenocopy of the K17 disorder pachyony- Supporting Information chia congenita type 2. Hum Mol Genet 1998; 7:1143–8. 9 Wilson NJ, Messenger AG, Leachman SA et al. Keratin K6c muta- Additional Supporting Information may be found in the online tions cause focal palmoplantar keratoderma. J Invest Dermatol 2010; version of this article at the publisher’s website: 130:425–9. Figure S1. Double-label immunofluorescence staining of 10 McLean WH, Hansen CD, Eliason MJ et al. The phenotypic and cells transiently transfected with K16 wild-type and mutant molecular genetic features of pachyonychia congenita. J Invest K16. (a) K16 wild-type, (b) K16 p.Arg418Cys and (c) K16 Dermatol 2011; 131:1015–7. 11 Moll R, Divo M, Langbein L. The human keratins: biology and p.Arg418Pro cDNAs in epithelial cell line PtK2. K16 was pathology. Histochem Cell Biol 2008; 129:705–33. detected with rabbit polyclonal antisera against human K16 12 Schweizer J, Bowden PE, Coulombe PA et al. New consensus and the endogenous K8 with monoclonal antibody LE41 to nomenclature for mammalian keratins. J Cell Biol 2006; 174:169– PtK2 K8. Nuclei were stained with 40,6-diamidino-2-phenylin- 74. dole (DAPI). In cells transfected with wild-type K16, 895% of 13 Steinert PM, Parry DA. Intermediate filaments: conformity and transfected cells showed a defined keratin cytoskeleton where diversity of expression and structure. Annu Rev Cell Biol 1985; 1:41– wild-type K16 co-localized with endogenous K8; the remain- 65. 14 Steinert PM, Yang JM, Bale SJ, Compton JG. Concurrence between ing 105% of transfected cells had filaments plus aggregates. In the molecular overlap regions in keratin intermediate filaments cells transfected with K16 p.Arg418Cys, similar to wild-type and the locations of keratin mutations in genodermatoses. Biochem K16, 89% of transfected cells had a normal filament network Biophys Res Commun 1993; 197:840–8. and 11% of cells had filaments plus aggregates. In contrast, in 15 Smith FJ, Liao H, Cassidy AJ et al. The genetic basis of pachyony- cells transfected with K16 p.Arg418Pro, 985% of cells con- – chia congenita. J Investig Dermatol Symp Proc 2005; 10:21 30. tained aggregates and there was collapse/aggregation of the 16 Wilson NJ, Leachman SA, Hansen CD et al. A large mutational – endogenous network; only 15% of transfected cells showed a study in pachyonychia congenita. J Invest Dermatol 2011; 131:1018 9 24. normal filament network. Original magnification 60. 17 Szeverenyi I, Cassidy AJ, Chung CW et al. The Human Intermediate Table S1. Polymerase chain reaction primers for specific Filament Database: comprehensive information on a gene family amplification of pachyonychia congenita-associated keratin involved in many human diseases. Hum Mutat 2008; 29:351–60. genes.

K17 N92S PC-K17 Data for IPCRR #515 N=116

Toenails Dystrophy 112 of 116 (97%) all 10 toenails thickened 87 of 116 (75%) 7-9 toenails thickened 10 of 116 (09%) 4-6 toenails thickened 7 of 116 (06%) 1-3 toenails thickened 8 of 116 (07%) Toenails - Onset N = 112 Birth or less than 1 year 79 of 112 (71%) 1 to 4 years old 24 of 112 (21%) 5 to 14 years old 9 of 112 (08%) 15 years and over 1 of 112 (01%) Fingernail Dystrophy 98 of 116 (84%) all 10 fingernails thickened 50 of 116 (43%) 7-9 fingernails thickened 12 of 116 (10%) 4-6 fingernails thickened 27 of 116 (23%) 1-3 fingernails thickened 9 of 116 (08%) Fingernails - Onset N = 98 Birth or less than 1 year 72 of 98 (73%) 1 to 4 years old 19 of 98 (19%) 5 to 14 years old 5 of 98 (05%) 15 years and over 3 of 98 (03%) Plantar Keratoderma 91 of 116 (78%) Always (never goes away) 74 of 116 (64%) Sometimes (clear up at times) 13 of 116 (11%) Seldom (feet usually clear) 4 of 116 (03%) Plantar Keratoderma- Onset N = 91 Birth or less than 1 year 12 of 91 (13%) 1 to 4 years old 29 of 91 (32%) 5 to 14 years old 37 of 91 (41%) 15 years and over 14 of 91 (15%) Plantar Pain 77 of 91 (85%) Often require meds for pain 17 of 91 (19%) Very painful, but do not use meds 28 of 91 (31%) Somewhat painful 32 of 91 (35%) Palmar Keratoderma 60 of 116 (52%) Always (never goes away) 16 of 116 (14%) Sometimes (clear up at times) 19 of 116 (16%) Seldom (hands usually clear) 24 of 116 (21%) Other Oral Leukokeratosis 29 of 116 (25%) Cysts 107 of 116 (92%) Follicular Hyperkeratosis 75 of 116 (65%) Natal or Prenatal Teeth 88 of 116 (76%)