Ophthalmic Manifestations of Xeroderma

Pigmentosum
A Perspective from the United Kingdom
Rongxuan Lim, BM BCh (Oxon) FRCOphth,1,2 Mieran Sethi, MBBS, BSc (Hons),1,3
Ana M.S. Morley, FRCOphth, MD1,2

Purpose: To document the ocular manifestations of xeroderma pigmentosum (XP), presenting via the United
Kingdom (UK) XP service, and to analyze the correlations between XP genotype and ophthalmic phenotype.
Design: Prospective observational case series.
Subjects: Eighty-nine patients seen by the UK Nationally Commissioned XP Service, from April 2010 to
December 2014, with a genetically confirmed diagnosis of XP.
Methods: Patients underwent a full ophthalmic examination at each visit. Clinical features from both eyes
were recorded on a standard proforma. The most recent assessments were analyzed. A 2-tailed Fisher exact test
was used to assess for differences in ocular features between patients in XP subgroups with impaired tran-
scription coupled nucleotide excision repair (TC-NER) (category 1: XP-A, B, D, F, and G) and preserved TC-NER
(category 2: XP-C, E, and V).
Main Outcome Measures: Lid and periocular abnormalities, ocular surface pathologies, neuro-
ophthalmologic abnormalities, lens and retinal abnormalities, and visual acuity (VA).
Results: Ninety-three percent of XP patients in our cohort had ocular involvement, with 65% describing
photophobia. The most common abnormalities were in the periocular skin and ocular surface, including inter-
palpebral conjunctival melanosis (44%) and conjunctival injection (43%). Eleven percent of patients had required
treatment for periocular cancers and 2% for ocular surface cancers. The most common neuro-ophthalmologic
finding was minimal pupillary reaction to light (25%). Patients in category 2 had significantly more ocular sur-
face abnormalities than patients in category 1, including a greater proportion of conjunctival injection (P ¼ 0.003),
conjunctival corkscrew vessels (P < 0.001), corneal scarring (P ¼ 0.01) and pingueculae under the age of 50
(P ¼ 0.02). Meanwhile, patients in category 1 had a higher proportion of poorly reactive pupils (P < 0.001) and
abnormal ocular movements (P ¼ 0.03) compared with those in category 2. Five patients (6%) presented to
ophthalmologists with ocular surface signs related to XP, before any formal diagnosis of XP was made.
Conclusions: A large proportion of XP patients have ocular involvement. Regular examination by an ophthal-
mologist is essential, especially in screening for eyelid and ocular surface tumors. The ocular phenotypeegenotype
segregation within XP patients suggests that XP is a heterogeneous and complex disease. With further study, we
hope to offer these patients more individualized patient care. Ophthalmology 2017;-:1e10 ª 2017 by the American
Academy of Ophthalmology

Xeroderma pigmentosum (XP) is a rare autosomal recessive
DNA repair disorder with ophthalmic, dermatologic, and
neurological manifestations.1,2 Most patients present to
dermatologists with abnormal freckling, extreme sunburn
reactions, or premature and multiple skin cancers.1,2 Pro-
gressive neurodegeneration occurs in 20% to 30% of cases,
initially affecting cognition, hearing, and mobility, but
which can also result in premature death.1,2 Ophthalmic
pathology has been described in 40% to 100% of XP pa-
tients and generally affects the sun-exposed periocular skin
and the ocular surface.3e6 Changes include ectropion,
lagophthalmos, conjunctival injection, conjunctival mela-
nosis, corneal scarring and keratopathy, pterygium, and
cancers of both the ocular surface and eyelids.3e6
Commonly reported ocular symptoms include photophobia
and dry eyes.5e7 There is also some early evidence that XP
patients experience an accelerated rate of corneal endothelial
cell loss.8
The estimated incidence of XP in the United States,
Japan, and Western Europe is 1 in 250 000,9 1 in 22 00010
and 1 in 500 000,11 respectively. The clinical
manifestations of XP are extremely varied and are
influenced by the precise genetic mutation (and therefore
XP complementation group), as well as environmental
factors, such as cumulative sun exposure.1,2
There are 8 known XP complementation groups (XP-A
to XP-G, and XP-V), which correspond to the 8 genes that
are affected in this condition. The proteins encoded by the
XPA to XPG genes are involved in nucleotide excision
repair (NER), which identifies and repairs DNA damage

ª 2017 by the American Academy of Ophthalmology http://dx.doi.org/10.1016/j.ophtha.2017.04.031 1

Published by Elsevier Inc. ISSN 0161-6420/17
 Ophthalmology Volume -, Number -, Month 2017
primarily induced by ultraviolet radiation (UVR).1,2,12 There
are 2 subpathways within NER: transcription-coupled
nucleotide excision repair (TC-NER), which repairs
actively transcribed DNA, and global genome nucleotide
excision repair (GG-NER), which repairs DNA not under-
going transcription at that time. XPC and XPE proteins only
contribute to GG-NER, whereas XPA, XPB, XPD, XPF,
and XPG proteins are required for both GG-NER and TC-
NER (Fig 1). XPV is not directly involved in NER; it
encodes a DNA polymerase (h) that is required for DNA
replication past damage that has not been repaired by
NER, in a process known as translesion synthesis.13,14
XP patients from complementation groups C, E, and V
(where TC-NER is preserved) have normal sunburn re-
actions for skin type and do not develop manifest neuro-
degeneration.15,16 However, they have an earlier age of
onset of first skin cancer.15 However, XP patients from
complementation groups A, B, D, F, and G (where TC-
NER is impaired) have severe and exaggerated sunburn
reactions on minimal sun exposure and suffer
neurodegeneration.15,16
There is currently no cure for XP. Educating patients about
the importance of rigorous photoprotection, as well as regular
screening for cancerous lesions, is paramount in the man-
agement of this condition.17 Due to the difficulties many
patients face in accessing the necessary range of hospital
services for screening and advice, a nationally
commissioned multidisciplinary XP service was set up
in the United Kingdom (UK) in 2010, with input
from dermatologists, neurologists, ophthalmologists,
psychologists, and geneticists.18 Patients are generally seen
once or twice a year and undergo a comprehensive review
by the multidisciplinary team in a single visit that usually
spans the day. In this study, we specifically describe the
ocular manifestations of our patients in the UK National XP
Service and the correlations between genotype and
ophthalmic phenotype that we have observed. We also
highlight the crucial role that ophthalmologists play in the
diagnosis and management of this condition.
Methods
We conducted a prospective observational case series of all pa-
tients seen by the UK National XP Service, from its establishment
in April 2010 until December 2014, in whom a genetic diagnosis of
XP had been confirmed. The study was approved by the Research
Ethics Committee of Guy’s and St Thomas’ Hospital National
Health Service (NHS) Trust, London (reference 12/LO/0325).
Informed consent was obtained from all patients and the study
adhered to the tenets of the Declaration of Helsinki.
At each clinic attendance, all patients were seen by a single
consultant ophthalmologist, or occasionally by a senior ophthalmic
colleague. A full ophthalmic examination was conducted at each
visit, including Snellen visual acuity (VA), Ishihara color vision,
Goldman tonometry for intraocular pressure (IOP), ocular motility,
pupillary responses, lid position and function, and slit-lamp
assessment of the ocular surface, anterior chamber, lens, and
fundus. All patients were also asked whether they experienced
photophobia. The findings were recorded on a standard proforma
each time and the clinical features of both eyes from each patient’s
most recent assessment were used in the data analysis.
2
The laboratory methods of XP genotyping used, as well as
amalgamated data on the dermatologic, neurological, and ophthal-
mologic findings of the cohort, has been described in a previous
article from our group19 and will not be repeated here. However, this
study will discuss the ophthalmologic findings, including periocular
and neuro-ophthalmic findings, in greater detail.
As mentioned in the introduction, various phenotypic differ-
ences have been reported between the 8 XP complementation
groups. These differences are most marked when comparing those
groups in which TC-NER is impaired (i.e., XP-A, B, D, F, and G)
with those where TC-NER is preserved (i.e., XP-C, E, and V)
(Fig 1). As such, we performed a statistical analysis on the
ophthalmic features of our patients by subdividing them into
these 2 categories, termed category 1 and category 2,
respectively. Statistical analysis was computed using SPSS v24.0
(IBM Corp, Armonk, NY) and a 2-tailed Fisher exact test was
always used unless otherwise specified.
Finally, we also used an overall quantitative score for
ophthalmic pathology to look at the specific ophthalmic morbidity
of individual complementation groups. This score was similar to
that described in our group’s previous article,19 but here we have
included neuro-ophthalmic, fundal, and periocular features as
part of the scoring.
Results
A total of 89 patients with a genetically confirmed diagnosis of XP
were assessed successively in the study period and all were included.
The XP complementation groups of these patients were as follows:
18 XP-A, 2 XP-B, 28 XP-C, 14 XP-D, 4 XP-E, 3 XP-F, 8 XP-G, and
12 XP-V. There were 45 female patients and 44 male patients. The
median age in December 2014 was 23 years (range, 4e80 years)
(Table 1), with notably older ages for patients in complementation
groups B, E, and V. The represented ethnicities, based on the
Office for National Statistics’ breakdown of ethnic categories for
classification from the 2001 census, are shown in Table 2. XP-A
and XP-C were largely composed of patients of Eastern origin,
including Pakistani and Bangladeshi patients, whereas XP-D, XP-E,
XP-F, and XP-V comprised mostly white patients.
Eighty-three of 89 (93%) XP patients in this study had at least 1
ophthalmic abnormality. For ease of analysis, ophthalmic features
were categorized into 1 of the following 4 groups: (1) periocular/
eyelid abnormalities; (2) ocular surface abnormalities; (3) neuro-
ophthalmic abnormalities; (4) lens/retinal abnormalities. A sum-
mary of the above results is given in Tables 3 and 4.
As VA, color vision, and IOP could not be reliably assessed in
all pediatric or neurologically impaired patients, results from these
ophthalmic assessments are discussed separately. The best-
corrected visual acuity (BCVA) of the better eye of each patient
is also shown in Table 5.
Eyelid Abnormalities
Fifteen of 89 (17%) patients had either ectropion or lagophthalmos
on gentle closure, but none had entropion. All ectropia were related
to cicatricial skin changes, principally secondary to surgery for
management of periocular skin cancers (Fig 2A), but also due to
dry, tight skin. Lagophthalmos was caused by upper lid scarring
from surgery (Fig 2B) or marked periocular fat atrophy resulting
in enophthalmos and superior sulcus hollowing. No statistically
significant difference was found in the prevalence of ectropion or
lagophthalmos between categories 1 and 2.
At the time of our assessment 10 of 89 (11%) patients had been
diagnosed with and treated for eyelid skin tumors; this included 2
squamous cell carcinomas (1 XP-C and 1 XP-V patient), 7 basal
 Lim et al 
Ophthalmic Manifestations of XP

Figure 1. Role of xeroderma pigmentosum (XP) proteins in the repair of ultraviolet radiation (UVR) damage (taken from Ref. 19). A, Nucleotide excision
repair (NER). The figure represents a scheme for NER, showing only the roles of the XP proteins. For simplicity, other proteins involved have not been
included. In global genome (GG)-NER, XPE and XPC are involved in damage recognition. Transcription factor IIH (TFIIH), containing XPB and XPD, are
subsequently recruited and open out the DNA structure in the vicinity of the damage. XPA verifies the damaged structure before dual incisions by XPF-
ERCC1 and XPG. After the damaged section has been removed, DNA polymerases and ligases fill in the gap. In transcription-coupled (TC)-NER, blockage
of RNA polymerase by the damage is the recognition signal. Proteins CSB and CSA, defective in Cockayne syndrome, are recruited and these in turn recruit
TFIIH to enable the rest of the NER process to proceed. B, Translesion synthesis. DNA replication, carried out by DNA polymerases d or ε, is blocked by a
UVR damage lesion, but frequently reinitiates beyond it. The resulting gap is filled in by DNA polymerase h, which is deficient in XP variants.

cell carcinomas (2 XP-C, 3 XP-D, 1 XP-E, and 1 XP-V patient),
and 3 malignant melanomas (1 XP-C and 2 XP-V patients). Two
patients had both an upper-eyelid basal cell carcinoma and a lower-
eyelid malignant melanoma. The median age of patients with
eyelid tumors as of December 2014 was 58 years (range, 6e71
years). All periocular tumors had been managed with surgical
excision and reconstruction. No statistically significant difference
was found in the prevalence of eyelid tumors between categories 1
and 2.
Ocular Surface
Photophobia was a very commonly reported symptom, described
by 58 of 89 patients (65%) and reported by patients in all

3
 Ophthalmology Volume -, Number -, Month 2017

Table 1. Demographics of Patients in the Study According to Xeroderma Pigmentosum Complementation Group

XP-A XP-B XP-C XP-D XP-E XP-F XP-G XP-V Total

Number of patients 18 2 28 14 4 3 8 12 89
Men 10 0 14 7 0 2 4 7 44
Women 8 2 14 7 4 1 4 5 45
Median age as of December 2014 (yrs) 23 45 14 24 56 23 19 55 23

XP ¼ xeroderma pigmentosum.

complementation groups except XP-B. It was not clearly linked to
abnormalities of the ocular surface, neuro-ophthalmologic system,
or retina (Table 3).
Conjunctival Abnormalities. Thirty-eight of 89 (43%) patients
had conjunctival injection. There was a greater proportion of pa-
tients with conjunctival injection in category 2 compared with
category 1 (59% vs. 27%; P ¼ 0.003). Many patients had received
topical anti-inflammatory drops for this, before review in the XP
clinic, with little effect.
Twenty of 89 (22%) patients were found to have conjunctival
corkscrew vessels. These are typical prominent and tortuous ves-
sels (Fig 2C). Again, these were more common in patients from
category 2 as compared with category 1 (43% vs. 2%; P < 0.001).
Thirty-nine of 89 (44%) patients had interpalpebral conjunctival
melanosis. This may partly reflect the racial conjunctival pigmen-
tation present in some dark-skinned individuals, who form a large
proportion of our XP cohort (Table 2). However, we have often
observed increased conjunctival pigmentation in patients who
also have abnormal XP-related skin freckling, suggesting that
this conjunctival melanosis could be XP-related. Nonetheless,
interpalpebral conjunctival melanosis in fair-skinned white patients
was also assessed separately and found to be lower, at 4 of 42
(10%). There was no statistically significant difference in general
conjunctival pigmentation between the 2 categories.
Thirty-six of 89 (40%) patients exhibited pterygia, spread
relatively evenly between both categories. Of the 74 patients under
the age of 50, 21 (28%) had pingueculae, although these showed a
higher prevalence in category 2 (42% vs. 17%; P ¼ 0.02).
Corneal Abnormalities. The most common corneal abnor-
mality was corneal staining, in the form of superficial punctate
keratopathy, indicative of dry eyes. This was present in 19 of 89
(21%) patients and did not show a statistically significant differ-
ence between the 2 categories.
Fifteen of 89 (17%) patients were found to have corneal scar-
ring (Fig 2C). In general, this was peripheral and associated with
other corneal surface changes, such as pterygia, exposure
keratopathy, and mild corneal neovascularization. However, 2
siblings with XP-G from a consanguineous marriage were seen
with congenital, diffuse, generalized corneal opacification. The
older of these 2 had undergone bilateral penetrating keratoplasties
as a child, as well as bilateral cataract extractions. A third child
with XP-G, from a nonconsanguineous marriage, presented with a
smaller area of central corneal opacification, without any other
peripheral corneal changes. Overall, we found a greater proportion
of category 2 patients with corneal scarring compared with cate-
gory 1 (27% vs. 7%; P ¼ 0.01).
None of our patients had corneal edema, although this has
previously been described in the literature elsewhere.20,21
Ocular Surface Cancers. Two of 89 (2%) patients had a history
of ocular surface cancers. These consisted of 1 squamous cell
carcinoma at the limbus (XP-A) (Fig 2D) and 1 limbal melanoma
(XP-B). The age of 1 of the patients as of December 2014 was 24;
the other patient died during the review period aged 55 years.
Five other patients were also noted to have cellular dysplasia,
without invasive disease, on ocular surface biopsy. The biopsies were
taken either by our team or by the referring hospital for any suspicious
ocular surface lesions. Cases included 2 patients with conjunctival
intraepithelial neoplasia (XP-A and XP-C), 1 with limbal intra-
epithelial neoplasia (XP-C) (Fig 2E), and 2 with conjunctival
intraepithelial melanocytic proliferation with atypia (XP-A and XP-C).
Neuro-ophthalmic Abnormalities
Twenty-two of 89 (25%) patients were found to have sluggish
pupils that showed minimal or no constriction to light, despite
reasonable VA. All of these patients were from category 1 and
none was from category 2 (49% vs. 0%; P < 0.001).
Seven of 89 (8%) patients had strabismus, which consisted of 5
exotropias (examples in Fig 2F and G) and 2 esotropias. All the
patients with exotropias were in category 1, whereas all the
patients with esotropias were in category 2. However, we did not
find a statistical difference in the distribution of esotropias and

Table 2. Ethnicity of Patients in the Study According to Xeroderma Pigmentosum Complementation Group, Based on the Office for
National Statistics’ Breakdown of Ethnic Categories for Classification from the 2001 Census

Ethnicity XP-A XP-B XP-C XP-D XP-E XP-F XP-G XP-V Total
White 1 1 7 12 3 3 5 10 42
Pakistani 11 0 11 0 1 0 1 0 24
Bangladeshi 2 0 5 0 0 0 0 0 7
Middle Eastern 0 0 3 0 0 0 0 1 4
Somalian 2 0 0 0 0 0 2 0 4
Mixed 0 1 0 1 0 0 0 1 3
Indian Asian 2 0 0 0 0 0 0 0 2
Chinese 0 0 0 1 0 0 0 0 1
Sri Lankan 0 0 1 0 0 0 0 0 1
Other Asian 0 0 1 0 0 0 0 0 1

XP ¼ xeroderma pigmentosum.

4
 Lim et al 
Ophthalmic Manifestations of XP

Table 3. Summary of Number of Patients with Specific Ophthalmic Findings, Organized According to Xeroderma Pigmentosum
Complementation Group

XP-A XP-B XP-C XP-D XP-E XP-F XP-G XP-V Total (%)
N ¼ 18 N¼2 N ¼ 28 N ¼ 14 N¼4 N¼3 N¼8 N ¼ 12 N ¼ 89
Ectropion or lagophthalmos 2 0 6 3 0 1 1 2 15 (17)
Entropion 0 0 0 0 0 0 0 0 0 (0)
Periocular cancers 0 0 3 3 1 0 0 3 10 (11)
Photophobia 9 0 20 12 3 2 3 9 58 (65)
Conjunctival injection 5 1 20 2 3 0 4 3 38 (43)
Conjunctival corkscrew vessels 0 0 16 0 1 0 1 2 20 (22)
Interpalpebral melanosis regardless of skin pigmentation 13 0 21 1 0 0 2 2 39 (44)
Interpalpebral pigmentation in white subjects (n ¼ 42) 0 0 1 1 0 0 1 1 4 (10)
Pterygium 9 1 14 7 3 0 1 1 36 (40)
Pinguecula under the age of 50 (n ¼ 74) 2 0 9 3 0 0 2 5 21 (28)
Corneal scar or neovascularization 1 0 8 0 2 0 2 2 15 (17)
Corneal staining (superficial punctate keratopathy) 2 2 9 2 0 1 2 1 19 (21)
Ocular surface cancer 1 1 0 0 0 0 0 0 2 (2)
Sluggish pupils 7 1 0 8 0 1 5 0 22 (25)
Strabismus (total) 1 0 1 3 0 0 1 1 7 (8)
Exotropia 1 0 0 3 0 0 1 0 5 (6)
Esotropia 0 0 1 0 0 0 0 1 2 (3)
Abnormal extraocular movement 2 0 0 2 0 0 2 0 6 (7)
Optic disc abnormality 0 0 0 0 0 0 0 0 0 (0)
Lens abnormality 0 0 0 1 0 0 2 1 4 (4)
Retinal abnormality 0 0 0 0 1 0 2 1 4 (4)
Average age-adjusted ocular score* 0.1 0.07 0.3 0.1 0.07 0.06 0.1 0.06 0.1

XP ¼ xeroderma pigmentosum.
*Average age-adjusted ocular score is derived from the addition of the scores in all the above columns divided by the number of patients and the mean age of
the patients in each complementation group.

exotropias between the categories, possibly due to the small
number of patients with strabismus in our cohort.
Six of 89 (7%) patients had abnormal extraocular movements.
This consisted of limitation of eye movements, slow or incomplete
saccades, and nystagmus. All these patients were from category 1
(13% vs. 0%; P ¼ 0.03).
None of the patients in our cohort had any optic disc
abnormality.
Lens/Retinal Abnormalities
Four of 89 (4%) patients had some cataract but only 1 was younger
than 50 years of age. This latter patient, with XP-G and who has
been mentioned previously, had undergone bilateral cataract sur-
gery, together with bilateral corneal grafts for corneal opacification,
by the age of 18. In this patient’s case, the cataracts, and corneal
opacification, were felt to have been congenital.
Four out of 89 (4%) of our patients had retinal abnormalities. These
consisted of age-related drusen in 1 patient with XP-E, panretinal
photocoagulation for diabetic retinopathy in another with XP-V, and a
second pair of siblings with XP-G in their late 30s with mild macular
retinal pigment epithelial changes. Despite the above features, these
XP-G siblings had good VA of at least 6/9 in both eyes, although the
elder sibling had reduced color vision of 12/17 on the Ishihara color
test. No patient exhibited peripheral pigmentary retinopathy.
Visual Acuity, Color Vision, and Intraocular
Pressure
VA, color vision, and IOP were assessed for all patients where
possible. However, an accurate assessment was often not feasible
in young children or patients with severe neurological problems.
The results for these are therefore not included in the main data set.
Instead, Table 5 shows the BCVA in the better eye of all patients
where an assessment could be made.
BCVA in the better eye was assessed to be at least 6/9 for 76 of
84 (90%) patients assessed. Of the 8 patients with recorded BCVAs
worse than 6/9 in their better eye, 2 patients were the young sib-
lings discussed previously with XP-G who had congenital corneal
opacification and, in the case of the older sibling, additional cata-
ract surgery (BCVA: 6/24 and 6/36).
The remaining 6 patients who had recorded BCVAs worse than
6/9 in their better eye had severe neurological impairment. Three of
these were children for whom assessment of VA was very chal-
lenging and difficult to interpret. None of them objected to oc-
clusion of either eye and their parents had no concerns about their
vision. Three were adults whose ability to comprehend and
perform VA tests was very limited. One was unable to use a
pinhole and may have had a refractive problem. The other 2 had
previously recorded VAs of 6/9 with refraction but subsequently
became unable to perform the tests. It is possible that there may
have been a genuine simultaneous reduction in VA with this
neurological defect; however, their families did not describe any
difficulty with vision per se.
Color vision was normal (score of 16/17 on the Ishihara color
test) in 75 of 80 (94%) patients. Three patients had abnormal color
vision due to known congenital color blindness. A further 2 pa-
tients had slightly reduced color vision at 15/17 and 12/17, with the
first test plates read correctly. The patient with a score of 12/17 was
the elder of a pair of siblings with XP-G who exhibited subtle
macular retinal pigment epithelial changes, discussed in the retinal
section above. However, these color vision test results must be
interpreted with caution, given the limited sample size and small
magnitude of the abnormality.

5
 Ophthalmology Volume -, Number -, Month 2017

Table 4. Comparison of Distribution of Ophthalmic Features between Xeroderma Pigmentosum Patients in Category 1 and Category 2

Category 1* Category 2*
No. of Cases (%) No. of cases (%)
N ¼ 45 N ¼ 44 P Valuey
Ectropion or lagophthalmos 7 (16) 8 (18) 0.8
Periocular cancers 3 (7) 7 (16) 0.2
Photophobia 26 (58) 32 (73) 0.2
Conjunctival injection 12 (27) 26 (59) 0.003
Conjunctival corkscrew vessels 1 (2) 19 (43) <0.001
Intepalpebral melanosis regardless of skin pigmentation 16 (36) 23 (52) 0.1
Interpalpebral pigmentation in white subjects (n ¼ 42) 2 (9) 2 (10) 1.0
Pterygium 18 (40) 18 (41) 1.0
Pinguecula under the age of 50 (n ¼ 74) 7 (17) 14 (42) 0.02
Corneal scar or neovascularization 3 (7) 12 (27) 0.01
Corneal staining (superficial punctate keratopathy) 9 (20) 10 (23) 0.8
Ocular surface cancer 2 (4) 0 (0) 0.5
Sluggish pupils 22 (49) 0 (0) <0.001
Strabismus (total) 5 (11) 2 (5) 0.4
Exotropia 5 (11) 0 (0) 0.06
Esotropia 0 (0) 2 (5) 0.2
Abnormal extraocular movement 6 (13) 0 (0) 0.03
Optic disc abnormality 0 (0) 0 (0) 1.0
Lens abnormality 3 (7) 1 (2) 0.6
Retinal abnormality 2 (4) 2 (5) 1.0

XP ¼ xeroderma pigmentosum.
*Category 1 includes all patients of complementation group XP-A, B, D, F, and G and category 2 includes all patients of complementation group XP-C, E,
and V.
y
P value using Fisher exact test (2-tailed) comparing complementation groups XP-A, B, D, F, G, and V with XP-C, E, and V. Statistically significant figures
(P < 0.05) appear in bold italics.

IOP was only measured in cases where this could be done
without causing too much anxiety to the patient. This meant that
most children and some adults with severe neurological impair-
ment did not get IOP measurements. All tested patients (n ¼ 79),
except 1 adult with known ocular hypertension, had IOPs less than
21 mmHg. The latter was an XP-B patient in her 50s who had
undergone enucleation of her right eye previously and whose IOP
in her only eye measured 24 mmHg.
Average Age-Adjusted Ocular Score
To compare the level of ophthalmic pathology among the indi-
vidual XP complementation groups, we derived an average age-
adjusted ocular score for each one. This was calculated by sum-
ming the number of patients with each ocular abnormality as listed
in Table 3, dividing the total figure by the number of patients in
each complementation group and then further dividing by the
mean age of the patients in that complementation group, as XP-
related ocular pathology accumulates over time. XP-C patients
had the highest score (0.3) of all the XP complementation groups,
suggesting a higher age-adjusted level of ophthalmic pathology.
Discussion
Our study found a very large proportion of XP patients
(93%) with ocular involvement. This is expected, given the
high exposure of the eye to UVR, as well as the role the eye
has within the neurological system, and underscores the
importance of regular ophthalmic review for XP patients.
Most of these changes were found in the periocular skin
and ocular surface, as documented in the current literature.3e7
The most common changes in our study were photophobia
(65%), interpalpebral conjunctival melanosis regardless of
skin pigmentation (44%), and conjunctival injection (43%).
Significantly, some patients required treatment for eyelid
tumors (11%) and ocular surface tumors (2%).

Table 5. Distribution of Snellen Best-Corrected Visual Acuity of the Better Eye of Xeroderma Pigmentosum Patients in whom Testing
Was Possible (N ¼ 84)

BCVA
6/6 <6/6 to 6/9 <6/9 to 6/12 <6/12 to 6/18 <6/18 to 6/24 <6/24 to 6/36 <6/36 to 6/60 <6/60
Number of patients 56 20 4 0 2 1 0 1

BCVA ¼ best-corrected visual acuity.

6
 Lim et al 
Ophthalmic Manifestations of XP

Figure 2. A, Patient with xeroderma pigmentosum (XP)-D showing inferior displacement (retraction) of right lateral lower lid and medial ectropion of left
lower lid, secondary to previous eyelid surgery for periocular cancer management. B, Patient with XP-V showing significant left cicatricial lagophthalmos on
gentle closure of eyes, secondary to previous surgical excision of left supra-brow squamous cell carcinoma. C, Anterior segment photograph of patient with
XP-C showing conjunctival corkscrew vessels and early limbal scarring. D, Anterior segment photograph of patient with XP-A showing medial limbal
invasive squamous cell carcinoma. E, Anterior segment photograph of patient with XP-C showing limbal conjunctival intraepithelial neoplasia grade 3.
F, Exotropia in a patient with XP-A. G, Exotropia in a patient with XP-D.

A proportion of our patients had neuro-ophthalmic ab-
normalities in the form of minimal pupillary constriction to
light (25%), strabismus (8%), and abnormal extraocular
movements (7%). Notably however, none of our patients
had any manifest optic atrophy. Although neuro-
degeneration is recognized in a significant proportion of XP
patients in the literature,1,2,4,6,16 reports of specific neuro-
ophthalmic changes, such as ours, remain limited.22
Most of the patients in our study had good visual func-
tion. The BCVA in the better eye was assessed to be at least
6/9 in 90% of the 84 tested patients and, in those with lower
BCVA, may have related to poor cerebral processing rather
than any ophthalmic abnormality per se. This percentage is
higher than that previously reported in the global literature
and may be due to earlier diagnosis (compared with some
other countries), the naturally lower ambient UVR levels in
the UK, easier access to health care, and higher compliance
with meticulous photoprotection in our patients. For com-
parison, in their review of 830 published cases, Kraemer
et al6 found that 12% of patients had what they defined as
“impaired vision.” More recently, a study based in India
by Goyal et al5 reported a VA of less than 6/36 in 50% of
their cases, and a second Saudi Arabian study by Alfawaz
et al7 reported a VA of 6/60 or less in 55% of their
patients. These latter 2 results are markedly different from
ours. Brooks et al,3 based in the United States, found that
13% of their patients had some degree of visual axis
impairment, whereas 5% had no light perception in at
least 1 eye.
Our study also revealed several associations between
ophthalmic phenotypes and complementation groups.
Firstly, patients in complementation groups XP-C, E, and V
(category 2) had significantly more ocular surface abnor-
malities compared with patients in complementation groups
XP-A, B, D, F, and G (category 1). This included a greater
proportion of conjunctival injection (P ¼ 0.003), conjunc-
tival corkscrew vessels (P < 0.001), corneal scarring (P ¼
0.01), and pingueculae under the age of 50 (P ¼ 0.02). As
previously discussed, patients in category 2 do not generally
suffer from the severe sunburn reactions on minimal sun
exposure that affect patients in category 1.15 It is therefore
possible that, before a diagnosis of XP, patients in
category 2 spend more time in the sun and develop more
cumulative ocular surface damage than those in category 1.
Within category 2, we found that patients with XP-C had
a higher average ocular pathology score compared with
those with XP-E or XP-V (0.3 vs. 0.07 or 0.06, respectively;
Table 3). Our group has previously reported that XP-E and
XP-V patients generally have more skin cancers compared
with the XP-C patients.19 Taken together, these findings
suggest that there are factors other than cumulative UVR
exposure that influence the extent of ocular pathology XP

7
 Ophthalmology Volume -, Number -, Month 2017

patients develop. They also lend weight to the idea that parents and showed unusual mild macular speckling and
mutations in the different XP genes confer slightly reduced color vision, with more pronounced retinal pigment
different effects in different organs, perhaps because of epithelium granularity on optical coherence tomography.
specific local factors. Interestingly, the macular change does bear some resem-
We observed that a higher proportion of patients in cate- blance to another XP-G case described in the literature.3
gory 1 had pupils that reacted minimally to light compared Finally, the girl, also from nonconsanguineous parents,
with the patients in category 2 (P < 0.001), and that not a showed an unusual central congenital corneal opacity.
single patient in category 2 showed this feature. The cause of Such ophthalmic pathologies are not typical of XP but
this pupillary abnormality remains unclear, as visual function may represent a phenotypic overlap of these XP-G pa-
is not significantly affected in our XP patient cohort; however, tients with other DNA repair disorders, as has previously
it may represent a neurodegenerative problem with the been described in the literature.23,24 Alternatively, in the
efferent reflex pathway, not dissimilar to an Adie’s tonic case of 1 of the sibling pairs, it may be due to additional
pupil. Furthermore, there was a higher proportion of patients concurrent recessive ophthalmic pathologies, related to
with abnormal extraocular movements in category 1 parental consanguinity. Recruitment of more patients to the
compared with category 2 (P ¼ 0.03). This correlates with XP clinic and exchange of information with other XP ser-
reported data on clinically apparent neurodegeneration in XP, vices throughout the world should facilitate a greater un-
which is not manifested in category 2.15,16 derstanding of these unusual phenotypes.
Brooks et al3 noted that patients with a tendency to severe During this study, it also became apparent that 5 of our
sunburn reactions were less likely to have ectropion and XP cohort (6%) initially presented to ophthalmologists with
conjunctival melanosis, but were more likely to develop ocular surface signs related to their XP, before any formal
neoplastic ocular surface lesions, compared with patients diagnosis of XP had been made. Three had red, photophobic
with normal sunburn reactions for skin type. Our present eyes, which were repeatedly diagnosed as allergic eye dis-
study did not find any significant difference in ectropion ease, whereas 2 had premature pterygia. By raising aware-
and lagophthalmos (P ¼ 0.8), conjunctival melanosis ness of XP in the ophthalmology community, we hope that
(P ¼ 0.1), or ocular surface neoplasia (P ¼ 0.5) in patients the diagnosis of XP will be considered earlier in patients
in the complementation groups linked to severe sunburn presenting with ocular surface changes in the presence of
reactions (category 1) compared with those in either abnormal skin freckling or severe sunburn reactions
complementation groups with normal sunburn reactions for on minimal sun exposure.
skin type (category 2). However, due to the rarity of the
disease, the numbers of each complementation group in
both studies are small. This, coupled with the considerable
phenotypical variation in these groups and possible
different levels of UVR protection between patients,
renders difficulty in direct comparison of both studies.
In our statistical analysis, we chose to split the patients
into 2 categories (XP-A, B, D, F, and G [category 1] vs.
XP-C, E, and V [category 2]), mainly because of the dif-
ference in TC-NER function (impaired in category 1; pre-
served in category 2). However, within these 2 categories,
the number of patients within each complementation group
was not the same; for example, of the 44 patients in category
2, 28 (63%) were XP-C. The results of our study may
therefore be unduly influenced by the complementation
groups with larger numbers of patients, such as XP-C,
XP-A, and XP-D. As more XP patients are diagnosed and
the numbers of each group increase, we will be able to
perform more detailed analyses of the individual comple-
mentation groups compared with the whole cohort, which
will refine the genotypeephenotype correlations further.
It is also worth mentioning some unusual ophthalmic
phenotypes that we found in our cohort, which have been
referred to in the “Results” section. Specifically, within the
XP-G group we had 2 sets of sibling pairs, together with an
additional girl, all of whom had atypical ophthalmic fea-
tures. The younger of these sets of siblings was from
consanguineous parents and both siblings showed congen-
ital corneal opacification. The elder of the 2 also had
congenital or infantile cataracts, requiring bilateral cataract
surgery and penetrating keratoplasties during childhood. Figure 3. Visor used for facial and ophthalmic ultraviolet radiation
The other set of siblings was from nonconsanguineous protection.

8
 Lim et al 
Ophthalmic Manifestations of XP
For those patients with a confirmed diagnosis of XP,
referral to an ophthalmologist is essential. This allows the
detection and treatment of any eye pathologies, as well as
the screening for eyelid and ocular surface tumors. Some of
the most important areas of management include the treat-
ment of dry eyes, watery eyes, lagophthalmos, ectropion,
pterygia, ocular surface cancers, eyelid cancers, strabismus,
and double vision. In addition, ophthalmologists play a
fundamental role in advising and facilitating adequate
ophthalmic UVR protection. The gold standard for this is a
full-face visor, which is 100% ultraviolet A and ultraviolet B
protective, as shown in Figure 3. Older patients, who often
find the visor difficult, should use UVR-blocking glasses,
which can be either transparent or with a dark visual tint (as
with normal sunglasses) to reduce photophobia. Ideally,
these should have large lenses and sidebars to minimize
UVR reaching the eye. A peaked or wide-brimmed hat is
also essential in reducing overhead UVR exposure.25
In summary, our study serves to highlight the high
ophthalmic morbidity of XP and the essential role of oph-
thalmologists in managing this condition. Documentation of
this condition in the ophthalmic literature remains limited
and awareness of XP needs to be raised among the
ophthalmic community. The ophthalmic features we have
documented appear to show some phenotypeegenotype
correlation within XP, particularly in relation to whether or
not TC-NER is preserved. This segregation reflects other
phenotypic features, including severe sunburn on minimal
sun exposure and the development of neuro-
degeneration.15,16 The data also suggest that patients with
XP-C have the highest prevalence of eye disease, in
particular ocular surface problems. As these patients have
normal sunburn reactions for skin type and do not develop
manifest neurodegeneration, they may present to ophthal-
mologists first and it is essential that this disorder be
recognized. Finally, with further delineation of genoty-
peephenotype associations, we hope to glean further in-
sights into the pathogenesis of the disease and its
manifestations and offer patients more individualized care.
Acknowledgments. We thank Professor Alan Lehmann (Pro-
fessor of Genetics, University of Sussex) for review of the manu-
script. The genetic diagnosis and complementation group
classification of the patients were kindly provided by Professor
Alan Lehmann and Dr Shehla Mohammed (consultant clinical
geneticist). We also acknowledge Dr Robert Sarkany (consultant
dermatologist), Dr Hiva Fassihi (consultant dermatologist), and
members of the UK National Multidisciplinary XP Clinic team for
their invaluable work in the XP service.
References
1. Lehmann AR, McGibbon D, Stefanini M. Xeroderma pig-
mentosum. Orphanet J Rare Dis. 2011;6:70.
2. DiGiovanna JJ, Kraemer KH. Shining a light on xeroderma
pigmentosum. J Invest Dermatol. 2012;132(3 Pt 2):785-796.
3. Brooks BP, Thompson AH, Bishop RJ, et al. Ocular mani-
festations of xeroderma pigmentosum: long-term follow-up
highlights the role of DNA repair in protection from sun
damage. Ophthalmology. 2013;120(7):1324-1336.
4. Anttinen A, Koulu L, Nikoskelainen E, et al. Neurological
symptoms and natural course of xeroderma pigmentosum.
Brain. 2008;131(Pt 8):1979-1989.
5. Goyal JL, Rao VA, Srinivasan R, Agrawal K. Oculocutaneous
manifestations in xeroderma pigmentosa. Br J Ophthalmol.
1994;78(4):295-297.
6. Kraemer KH, Lee MM, Scotto J. Xeroderma pigmentosum.
Cutaneous, ocular, and neurologic abnormalities in 830 pub-
lished cases. Arch Dermatol. 1987;123(2):241-250.
7. Alfawaz AM, Al-Hussain HM. Ocular manifestations of
xeroderma pigmentosum at a tertiary eye care center in
Saudi Arabia. Ophthal Plast Reconstr Surg. 2011;27(6):
401-404.
8. Mohamed A, Peguda R, Ramappa M, et al. Corneal endo-
thelium in xeroderma pigmentosum: clinical specular micro-
scopy study. Br J Ophthalmol. 2016;100(6):750-753.
9. Robbins JH, Kraemer KH, Lutzner MA, et al. Xeroderma
pigmentosum. An inherited disease with sun sensitivity, mul-
tiple cutaneous neoplasms, and abnormal DNA repair. Ann
Intern Med. 1974;80(2):221-248.
10. Hirai Y, Kodama Y, Moriwaki S, et al. Heterozygous in-
dividuals bearing a founder mutation in the XPA DNA repair
gene comprise nearly 1% of the Japanese population. Mutat
Res. 2006;601(1-2):171-178.
11. Kleijer WJ, Laugel V, Berneburg M, et al. Incidence of DNA
repair deficiency disorders in western Europe: Xeroderma
pigmentosum, Cockayne syndrome and trichothiodystrophy.
DNA Repair. 2008;7(5):744-750.
12. Lehmann AR. DNA repair-deficient diseases, xeroderma pig-
mentosum, Cockayne syndrome and trichothiodystrophy.
Biochimie. 2003;85(11):1101-1111.
13. Lehmann AR, Kirk-Bell S, Arlett CF, et al. Xeroderma pig-
mentosum cells with normal levels of excision repair have a
defect in DNA synthesis after UV-irradiation. Proc Natl Acad
Sci U S A. 1975;72(1):219-223.
14. Masutani C, Kusumoto R, Yamada A, et al. The XPV
(xeroderma pigmentosum variant) gene encodes human DNA
polymerase eta. Nature. 1999;399(6737):700-704.
15. Sethi M, Lehmann AR, Fawcett H, et al. Patients with xero-
derma pigmentosum complementation groups C, E and V do
not have abnormal sunburn reactions. Br J Dermatol.
2013;169(6):1279-1287.
16. Bradford PT, Goldstein AM, Tamura D, et al. Cancer and
neurologic degeneration in xeroderma pigmentosum: long
term follow-up characterises the role of DNA repair. J Med
Genet. 2011;48(3):168-176.
17. Lambert WC, Lambert MW. Development of effective skin
cancer treatment and prevention in xeroderma pigmentosum.
Photochem Photobiol. 2015;91(2):475-483.
18. Lehmann AR. Xeroderma pigmentosum in the United
Kingdom. Photochem Photobiol. 2015;91(2):484-485.
19. Fassihi H, Sethi M, Fawcett H, et al. Deep phenotyping of 89
xeroderma pigmentosum patients reveals unexpected hetero-
geneity dependent on the precise molecular defect. Proc Natl
Acad Sci U S A. 2016;113(9):E1236-E1245.
20. Chaurasia S, Mulay K, Ramappa M, et al. Corneal changes in
xeroderma pigmentosum: a clinicopathologic report. Am J
Ophthalmol. 2014;157(2):495-500.
21. Chaurasia S, Ramappa M. Surgical intervention for corneal
involvement in xeroderma pigmentosa. Int Ophthalmol.
2012;32(1):1-2.
22. Ramkumar HL, Brooks BP, Cao X, et al. Ophthalmic mani-
festations and histopathology of xeroderma pigmentosum: two
clinicopathological cases and a review of the literature. Surv
Ophthalmol. 2011;56(4):348-361.
9
 Ophthalmology Volume -, Number -, Month 2017
23. Narita T, Narita K, Takedachi A, et al. Regulation of tran-
scription elongation by the XPG-TFIIH complex is implicated in
Cockayne syndrome. Mol Cell Biol. 2015;35(18):3178-3188.
24. Kraemer KH, Patronas NJ, Schiffmann R, et al. Xeroderma
pigmentosum, trichothiodystrophy and Cockayne syndrome: a
Footnotes and Financial Disclosures
Originally received: January 28, 2017.
Final revision: April 25, 2017.
Accepted: April 27, 2017.
Available online: ---. Manuscript no. 2017-93.
1
Nationally Commissioned Xeroderma Pigmentosum Service, Guy’s and
St Thomas’ NHS Foundation Trust, London, United Kingdom.
2 Data collection: Lim, Sethi, Morley
Department of Ophthalmology, Guy’s and St Thomas’ NHS Foundation
Trust, London, United Kingdom.
3
King’s College London, Kings Health Partners, Division of Genetics and
Molecular Medicine, St John’s Institute of Dermatology, Guy’s Hospital,
London, United Kingdom.
Presented at the Royal College of Ophthalmologists Annual Congress, May
21-23, 2013, Liverpool, United Kingdom; International Symposium on
Xeroderma Pigmentosum & Related Diseases, March 5-7, 2014, Kobe,
Japan; and European Xeroderma Pigmentosum Society Meeting, September
28, 2016, Vienna, Austria.
Financial Disclosure(s):
The National Multidisciplinary XP Clinic in the UK is funded by NHS
England Highly Specialised Services. Supported by the Medical Research
Council (fellowship grant no. MR/M001210/1 to M.S.), the UK National
Institute for Health Research (NIHR) Biomedical Research Centre (based at
Guy’s and St Thomas’ NHS Foundation Trust), and King’s College
10
complex genotype-phenotype relationship. Neuroscience.
2007;145(4):1388-1396.
25. Lim R, Fedele F, Patel P, Morley AM. Ocular solar protection in
xeroderma pigmentosum: the role of untinted lenses in blocking
ultraviolet radiation. Br J Dermatol. 2016;175(3):625-627.
London, UK. The sponsor or funding organization had no role in the design
or conduct of this research.
Author Contributions:
Conception and design: Lim, Morley
Analysis and interpretation: Lim, Sethi, Morley
Obtained funding: Not applicable
Overall responsibility: Lim, Sethi, Morley
Abbreviations and Acronyms:
BCVA ¼ best-corrected visual acuity; IOP ¼ intraocular pressure; GG-
NER ¼ global genome nucleotide excision repair; NER ¼ nucleotide
excision repair; NHS ¼ National Health Service; TC-NER ¼ transcription-
coupled nucleotide excision repair; UK ¼ United Kingdom;
UVR ¼ ultraviolet radiation; VA ¼ visual acuity; XP ¼ xeroderma
pigmentosum.
Correspondence:
Rongxuan Lim, BM BCh (Oxon) FRCOphth, Nationally Commissioned
Xeroderma Pigmentosum Service, Guy’s and St Thomas’ NHS Foundation
Trust, London SE1 7EH, United Kingdom. E-mail: limrongxuan@gmail.
com.