Incidence and profile of skin cancers

in patients following ultraviolet

phototherapy without psoralens:
A retrospective cohort study
Elle Wang, MSc,a,b Tashmeeta Ahad, MD,a,b Yi. A. Liu, MD,c Tim K. Lee, PhD,a,b,d Harvey Lui, MD,a,b
Richard I. Crawford, MD,c and Sunil Kalia, MDa,b,d,e

Background: Psoralen 1 ultraviolet-A (PUVA) is associated with photocarcinogenesis. However,

carcinogenic risk with other ultraviolet phototherapies remains unclear.

Objective: Evaluate whether phototherapy without psoralens increases skin cancer risk.

Methods: Retrospective cohort study of patients treated at a teaching-hospital phototherapy center (1977-
2018). Skin cancer records were validated against pathology reports. Age-standardized incidence rates
(ASIRs) of skin cancer were evaluated for gender, skin phototype, diagnosis, ultraviolet modality,
anatomical site; and compared to provincial population incidence rates (2003).

Results: In total, 3506 patients treated with broadband-ultraviolet-B, narrowband-UVB and/or combined
UVAB were assessed with a mean follow-up of 7.3 years. Majority of patients had psoriasis (60.9%) or
eczema (26.4%). Median number of treatments was 43 (1-3598). Overall, 170 skin cancers (17 melanoma, 33
squamous cell carcinoma and 120 basal cell carcinoma) occurred in 79 patients. Patient-based and tumor-
based ASIR of skin cancer was 149 (95% CI: 112-187)/100,000 and 264 (219-309)/100,000 person-years,
respectively. There was no significant difference between tumor-based ASIRs for melanoma, squamous cell
carcinoma, and basal cell carcinoma compared to the general population; or in phototherapy patients with-
psoriasis or eczema; or immunosuppressants. No cumulative dose-response correlation between UVB and
skin cancer was seen.

Limitations: Treatment and follow-up duration.

Conclusion: No increased risk of melanoma and keratinocyte cancer was found with phototherapy. ( J Am
Acad Dermatol 2024;90:759-66.)

From the Department of Dermatology and Skin Science, University University of British Columbia Hospital Foundation and Michael
of British Columbia, Vancouver, British Columbia, Canadaa; Smith Foundation for Health Research [18609]. The funders had
Photomedicine Institute, Vancouver Coastal Health Research no role in study design, data collection and analysis, decision to
Institute, Vancouver, British Columbia, Canadab; Department of publish, or preparation of the manuscript.
Pathology and Laboratory Medicine, University of British Patient consent: Not applicable.
Columbia, Vancouver, British Columbia, Canadac; Department IRB approval status: This study was reviewed and approved by the
of Cancer Control Research, BC Cancer, Vancouver, British University of British Columbia Clinical Research Ethics Board
Columbia, Canadad; and BC Children’s Hospital Research (CREB #H17-03410).
Institute, Vancouver, British Columbia, Canada.e Accepted for publication November 22, 2023.
Ms Wang and Dr Ahad are joint first authors. Correspondence to: Sunil Kalia, MD, Department of Dermatology
Funding sources: This study was supported by research grants and Skin Science, University of British Columbia, 835 W 10th
from the Eczema Society of Canada [F16-05485] and Canadian Ave, Vancouver, British Columbia V5Z 4E8, Canada. E-mail:
Dermatology Foundation. Dr Ahad has had funding from- The sunil.kalia@vch.ca.
Geoffrey Dowling Fellowship, British Association of Dermatol- Published online December 7, 2023.
ogists; Mr Lindsay Hall donation to the University of British 0190-9622/$36.00
Columbia; Canadian Dermatology Foundation and Canadian Ó 2023 by the American Academy of Dermatology, Inc.
Institute of Health Research (CIHR-IMHA). Dr Kalia is funded by https://doi.org/10.1016/j.jaad.2023.11.053
the Photomedicine Institute, Vancouver General Hospital and

759
760 Wang et al J AM ACAD DERMATOL
APRIL 2024

Key words: basal cell carcinoma; broadband; incidence; keratinocyte cancer; melanoma; narrowband;
phototherapy; skin cancer; squamous cell carcinoma; ultraviolet-B; ultraviolet light therapy; UVA; UVB.

INTRODUCTION receiving whole-body PUVA were excluded.8,23

Ultraviolet phototherapy Patients with vitiligo were
is commonly used for multi- treated at a separate unit and
ple skin conditions CAPSULE SUMMARY therefore not included. The
including psoriasis, atopic study was approved by the
eczema, mycosis fungoides, d
The risk of skin cancer with ultraviolet University of British
and pruritus.1-6 Since expo- phototherapy without psoralens has Columbia Clinical Research
sure to ultraviolet radiation been unclear Ethics Board (CREB-#H17-
from sunlight is strongly 03410).
d This study shows no significant
associated with skin cancer, Data abstracted manually
difference in incidence rates of
this risk is a consideration from patient records included
melanoma, SCC and BCC between the
when recommending pho- demographics, indication for
general population and patients treated
totherapy.7 The increased phototherapy, diagnosis, skin
with broadband-UVB, narrowband-UVB
risk of skin cancer with psor- cancer history, phototherapy
and combined UVB 1 UVA. No
alen 1 ultraviolet-A (PUVA) parameters (UV source, indi-
cumulative dose-response between UVB
has already been estab- vidual doses, dates). Skin can-
and skin cancer was seen
lished8; however, whether cer records for basal cell
there is an association with carcinoma (BCC), squamous
ultraviolet phototherapy cell carcinoma (SCC), and
without psoralens remains unsettled.9-23 melanoma were obtained and validated with pathol-
Photocarcinogenic risk may also vary by indication. ogy reports from a multi-institutional provincial
Studies have attempted to assess the risk of photo- electronic database, covering the majority of the
carcinogenesis with narrowband UVB (NB-UVB) in population. All skin cancers occurring after initiation
patients with psoriasis and vitiligo. However, results of phototherapy were included. If multiple skin
have been inconsistent, although most suggest no cancers occurred in a patient, these were analyzed
increased risk.9-23 Studies tend to be constrained by as separate skin cancers. Recurrent skin cancers were
low patient numbers, limited follow-up, lack of excluded.
appropriate comparator groups, and underestimation
due to not accounting for multiple skin cancer Descriptive analysis
occurrences, insufficient information regarding pho- Censored age was calculated by subtracting birth
totherapy dosimetry, and confounding factors such as year from end of follow-up year. Follow-up (person-
immunosuppressive medications. This study ad- years) was determined from date of phototherapy
dresses these limitations to more thoroughly investi- initiation to study follow-up end date (June 14, 2019).
gate skin cancer risk in a large cohort of patients Cumulative fluences and number of treatments were
treated between 1977 and 2018; by evaluating patient- calculated for each UV source. Skin cancer incidences
based and tumor-based incidences of skin cancer, were categorized by age, gender, skin phototype,
validating these with pathology reports, correlating anatomical site, and previous skin cancer history.
with number of exposures, cumulative ultraviolet
dose, disease, immunosuppressive medications; and Skin cancer incidence
characterizing profile of skin cancers by type, location, Both per patient and per tumor incidences of skin
and patient demographics. cancers were tracked to account for individual
patients who developed multiple skin cancers.
METHODS Patient-based incidences (number of patients with
A retrospective cohort study of patients treated at first occurrence of skin cancer after phototherapy)
the phototherapy center, Vancouver General and tumor-based incidences (total number of new
Hospital, Canada was undertaken. Patients were skin cancers) are expressed per 100,000 person-
treated with NB-UVB (311-313 nm), broadband years with 95% CI.
UVB (BB-UVB) (280-320 nm) and combined UVAB Age-standardized incidence rates (ASIR) of skin
(280-400 nm). Chart records of patients undergoing cancers with 95% CI have been calculated by age-
whole-body ultraviolet phototherapy from May matching to the 1991 Canadian population.24 This
1977-November 2018 were reviewed. Patients enabled comparison of patient-based and tumor-
J AM ACAD DERMATOL Wang et al 761
VOLUME 90, NUMBER 4

RESULTS
Abbreviations used:
Study population
ASIR: age-standardized incidence rate Records for 3554 patients were reviewed; 14
BB-UVB: broadband UVB
BCC: basal cell carcinoma patients who did not receive phototherapy and 34
KC: keratinocyte cancer patients who had previous systemic PUVA were
NB-UVB: narrowband UVB excluded, leaving 3506 participants for analysis.
PUVA: psoralen 1 ultraviolet-A
SCC: squamous cell carcinoma Median age at end of follow-up was 50 (range
UVA: ultraviolet-A 16-100). 57.5% (n = 1999) patients were male. The
UVB: ultraviolet-B age range at which patients started phototherapy
was 7-93 years; 28 patients started phototherapy in
childhood (age \18). Patients had Fitzpatrick skin
based (primary outcome) skin cancer ASIRs between phototypes I to VI (Table I). The majority of patients
our cohort and the 2003 British Columbia provincial had psoriasis (60.9%; n = 2136) or eczema (26.4%;
population, previously reported by Mclean et al, also n = 925); 2.3% (n = 82) of patients had a history of
age-standardized to 1991 Canadian census data.25 prior skin malignancy.
The British Columbia Cancer Registry codes for
total number of melanomas, including multiple
melanomas in individual patients (ie tumor-based Phototherapy source, number of sessions, and
incidence). The presence of multiple BCCs or SCCs cumulative dose
in individual patients is identified and coded sepa- Patients were exposed to various UV modalities
rately in this registry. However, only second primary either as the sole source or in combination (Fig 1).
BCCs or SCCs with unspecified sites are included in The median number of exposures for any UVB in
its ASIR calculations and therefore, not all multiple aggregate was 42 (range 1-3598); and for BB-UVB,
keratinocyte cancers (KCs) will have been included NB-UVB, and UVAB were 35 (1-3328), 43 (1-1487),
in the provincial reference population data. Thus, and 25 (1-881) respectively. The median cumulative
because of this particular ascertainment approach for dose for BB-UVB was 3.6 (0.01-1502)J/cm2 and NB-
patients with multiple KCs within this registry, the UVB 18.9 (0.01-2466)J/cm2. For UVAB treatment, the
provincial population KC ASIR is expected to be median UVA and UVB cumulative fluences were
possibly lower than the actual population tumor- 101.5 (1-8654) and 1.3 (0.02-197.3)J/cm2 respec-
based incidence. tively; several patients (n = 998 [28.5%]) received
[100 sessions.
Incidence of skin cancer
Correlation between number of treatments, Mean overall follow-up for patients undergoing
cumulative ultraviolet dose, and skin cancer phototherapy was 7.3 (standard deviation 5.9; range
Patients completing phototherapy before 2011 had 0.6-42; median 5.9) years. For patients receiving BB-
their treatment records moved for storage to an UVB or NB-UVB specifically, the median follow-up
alternate location and were unavailable for review. was 7.2 and 4.0 years, respectively.
To avoid analytical bias, this particular analysis was Overall, 170 tumors (17 melanomas, 33 SCCs
performed in patients starting treatment after 2011 and 120 BCCs) were identified in 79 patients
(2650 patients). (Supplementary Table I, available via Mendeley at
To avoid error resulting from analyzing correlated https://doi.org/10.17632/b49j2ww9vy.1). The mean
repeated events, time to developing the first skin censored age of patients with skin cancer was 66 (28-
cancer was used for this analysis. Total treatment 95). Fifteen of these patients had a history of skin
sessions and cumulative fluences were grouped into cancer prior to initiating phototherapy. Two separate
3 ordinal terciles (ie, low, mid, and high). Multivariate patients were on concurrent immunosuppressive
logistic regression and Cox-regression analyses were medications (methotrexate and cyclosporine respec-
used to assess the effects of total number of exposures tively). Skin phototype profiles of these 79 patients
and cumulative fluence on skin cancer risk, and odds were: skin type I (n = 4), II (n = 24), III (n = 34), IV
and hazard ratios calculated. Survival analysis was used (n = 10), and V (n = 2).
to depict skin cancer development during the follow- Multiple skin cancers occurred in 27 patients.
up period. Nine patients had a prior history of skin cancer. One
The z test was used to test differences between separate patient was the individual taking cyclo-
incidence rates. Statistical analyses were performed sporine. In addition, 2 patients received UVAB
with R-Foundation for Statistical Computing (Austria, phototherapy. Although patients with previous skin
Version-3.6.1, 2019). cancer had higher ASIRs, probability of developing
762 Wang et al J AM ACAD DERMATOL
APRIL 2024

Table I. Patient demographics and diagnoses in

patients receiving ultraviolet phototherapy
(n = 3506)
Number of patients Proportion of
Characteristics (n = 3506) patients (%)
Gender
Male 1999 57.0
Female 1507 43.0
Skin phototype
I 83 2.4
II 566 16.1
III 1395 39.8
IV 917 26.2 Fig 1. Distribution of modalities used by 3506 patients
V 475 13.5 receiving one or more different forms of ultraviolet
VI 44 1.3 phototherapy. The number after each modality label
Unknown 26 0.7 indicates the total number of patients who received that
Diagnosis form of phototherapy. The numbers within the over-
Psoriasis 2136 60.9 lapping Euler diagram ellipses indicate the numbers (and
Eczema 925 26.4 percentages) of patients receiving single versus multiple
Pruritus 211 6.0 sequential phototherapy modalities. BB-UVB, Broadband
Lichen planus 52 1.5 ultraviolet B; NB-UVB, narrowband ultraviolet B; UVAB,
Mycosis fungoides 45 1.3 combined ultraviolet A plus BB-UVB.
Polymorphous 40 1.1
light eruption
To compare incidence rates to the general popu-
Granuloma annulare 24 0.7
Other* 73 2.1 lation, tumor-based ASIRs for melanoma, SCC and
Previous skin cancer BCC were compared to the 2003 British Columbia
(before treatment) provincial population, which reported ASIRs by
Yes 82 2.3 gender (Fig 2). The specified provincial KC ASIR is
No 3424 97.7 likely to be lower than the actual population inci-
dence as not all multiple KCs are included in the
*Other: granuloma annulare; progressive macular hypomelanosis; population data. Nevertheless, despite this,
folliculitis; eosinophilic folliculitis; pityriasis rosea; parapsoriasis;
solar urticaria; lymphomatoid papulosis; morphea; mastocytosis;
no significant difference was found between
delusions of parasitosis; lichen sclerosus; Grover’s disease; acne; melanoma, SCC (P [.05) and BCC (P [.05) tumor-
follicular mucinosis; erythroderma; cutaneous plasmacytoma; based ASIRs between patients undergoing photo-
lichen amyloidosis; and pigmented purpuric dermatosis. Patients therapy and the provincial population. When
with vitiligo were treated at a separate unit and therefore not comparing patient-based instead of tumor-based
included.
ASIRs to the general population, male patients
showed a significantly lower incidence of BCCs
subsequent skin cancers was not higher than similar (P \.05).
patients reported (Supplementary Table II, available
via Mendeley at https://doi.org/10.17632/
b49j2ww9vy.1). Incidence of skin cancer in phototherapy
ASIR of skin cancer have been assessed for cancer patients according to skin phototype, gender,
type, anatomical site, gender and UV modality anatomical site
(Supplementary Table III, available via Mendeley at The overall crude patient incidence of skin cancer
https://doi.org/10.17632/b49j2ww9vy.1). was 317 (95% CI: 253-394)/100,000 person-years,
The overall patient-based ASIR for skin cancer increasing with age from 54 (14-147) for patients
was 149 (95% CI: 112-187)/100,000 person-years. under age 40 to 602 (461-772) per 100,000 person-
Patient-based ASIRs for melanoma, SCC and BCC years above age 60 (Supplementary Fig 1A, available
were 24/100,000, 34/100,000 and 93/100,000 person- via Mendeley at https://doi.org/10.17632/b49j2w
years, respectively. w9vy.1).
Overall tumor-based ASIR was 264 (95% CI: 219- There was no significant difference in ASIRs
309)/100,000 person-years. Tumor-based ASIRs for of skin cancer between males and females
melanoma, SCC and BCC were 35/100,000, 48/ (Supplementary Fig 1B, available via Mendeley
100,000 and 182/100,000 person-years, respectively. at https://doi.org/10.17632/b49j2ww9vy.1).
J AM ACAD DERMATOL Wang et al 763
VOLUME 90, NUMBER 4

Comparison of age-standardized skin cancer incidence

UVB phototherapy patients BC general population

Age-standardized incidence (tumors/100,000 person-years)
220

200

180

160

140

120

100
186 189
80
140
132
60

40
56 53
20 39 33
22 29
17 13
0
MM BCC SCC MM BCC SCC

Males Females

Fig 2. Comparison of age-standardized incidence rates of skin cancer between patients

undergoing ultraviolet phototherapy and the general population (British Columbia [BC]
provincial population).

The majority of skin cancers (50.6% [n = 86]) were

located on the head and neck, followed by trunk
(25.3%), upper limb (12.9%), and lower limb (10.6%)
(Supplementary Figs 1D and 2, available via Mendeley
at https://doi.org/10.17632/b49j2ww9vy.1). Truncal
sites had the highest incidence of melanoma and facial
sites had the highest incidence of KCs.

Skin cancer in patients with psoriasis versus

eczema treated with phototherapy
Compared to patients with psoriasis, those with
eczema received fewer UVB treatment sessions.
There was no significant difference in patient-based
and tumor-based ASIRs of skin cancer between pso-
riasis and eczema patients treated with phototherapy
Fig 3. Cumulative occurrence of skin cancer over time (Supplementary Table IV, available via Mendeley at
following initiation of UV phototherapy. There were no https://doi.org/10.17632/b49j2ww9vy.1). A detailed
significant differences in skin cancer development analysis of skin cancer risk in our eczema patient
amongst patients treated with low, medium and high cohort has been published.26
number of treatment sessions with all UVB phototherapy
modalities (data not shown).
Skin cancer in phototherapy patients with a
history of immunosuppressive medications
Overall patient-based skin cancer ASIRs Patients with a history of immunosuppressive
decreased with higher skin phototypes but this was medications may be predisposed to skin cancer. A
not statistically significant (Supplementary Fig 1C, subgroup analysis was conducted to assess this on
available via Mendeley at https://doi.org/10.17632/ 2269 (65%) patients for whom complete medication
b49j2ww9vy.1). Skin phototype data are currently history was available; of which, 530 (23%) had a
not recorded in cancer registries. history of immunosuppressants. There was no
764 Wang et al J AM ACAD DERMATOL
APRIL 2024

significant increased risk of skin cancer in these keratoses in patients treated with more than 200
patients compared to patients with no immunosup- NB-UVB sessions.14 A recent Singaporean study
pressive medications (odds ratio: 0.96 [95% CI 0.4-2.0] reported possible increased risk of KC with photo-
[P = .54]) (Supplementary Table V, available via therapy, although follow-up was short with low
Mendeley at https://doi.org/10.17632/b49j2ww9vy. tumor numbers.28 A Taiwanese population study
1); suggesting this was not a significant confounding found no increased hazard ratio for development of
factor in this study. KC or melanoma in patients with uremic pruritus
receiving UVB phototherapy.29
Correlation between number of exposures, Previous studies have also suggested that patients
cumulative ultraviolet fluence, and skin cancer with atopic eczema and psoriasis may have a higher
Overall, 1302 patients received #25 sessions and risk of malignancy, including KC, melanoma and
998 patients [100 sessions. There was no significant lymphoma, although results have been inconsis-
difference in patient-based ASIRs between patients tent.30-40 In our recent study investigating skin cancer
treated with #25 compared to [100 exposures risk in patients with eczema treated with photo-
(Supplementary Table VI, available via Mendeley at therapy26 and in this larger cohort of patients
https://doi.org/10.17632/b49j2ww9vy.1). For the including other dermatoses, no increased risk of
survival analysis, 2650 patients followed up for a skin cancer was found in patients treated with
median of 5.1 (range: 16-100) years were included; phototherapy.
29 patients developed skin cancer (2 melanomas, 7 Overall, there were 170 skin cancers (17 mela-
SCCs, 20 BCCs). Overall, there was no significant nomas, 33 SCCs, and 120 BCCs) identified in 79
correlation between the total number of treatments patients post phototherapy. The anatomical distri-
or cumulative UV dosage and skin cancer for both bution of skin cancers was mostly on the head and
multivariate logistic regression and Cox-regression neck as per the general population. Skin cancer
models. Cumulative cancer curves show no signifi- incidence increased with age and decreased with
cant risk differences between high versus low tercile higher Fitzpatrick skin phototypes. These factors
groups for both number of exposures and cumula- suggest that the profile of skin cancers seen in
tive UV received for NB-UVB and BB-UVB (Fig 3, phototherapy patients is similar to the general pop-
Supplementary Fig3, available via Mendeley at ulation, consistent with natural sun exposure as a
https://doi.org/10.17632/b49j2ww9vy.1). primary causal factor.
We also evaluated skin cancer risk based on total
DISCUSSION number of sessions and cumulative dosage of UV
Overall, in this study of 3506 patients with psori- received. There was no significant difference in
asis, eczema, and other dermatoses treated with BB- ASIRs between patients treated with high versus
UVB, NB-UVB, and combined UVAB, we found no low number of treatments (#25 vs [100 sessions).
increased risk of melanoma, SCC, or BCC when No dose-response correlation was seen between
compared to the general population. Patients had a cumulative fluence or number of treatments and
mean follow-up of 7.3 years. development of skin cancer (Fig 3, Supplementary
Although ultraviolet phototherapy has been uti- Fig 3, available via Mendeley at https://doi.org/10.
lized for several years, skin cancer risk with pro- 17632/b49j2ww9vy.1). There is currently no clear
longed treatment is unclear due to difficulty in consensus on maximum lifetime limits for photo-
performing long observational studies to assess therapy with the exception of systemic PUVA.41
carcinogenic risk adequately. Diffey and Farr have Hearn et al found no significant difference in
suggested that accurate assessment of carcinogenic incidence rate ratios between patients receiving
risk with NB-UVB may require large multicenter \25 vs $100 sessions, although tumor numbers
studies and follow-up periods of more than were small.13 Other studies have shown similar
10 years.27 A systematic review of 4 studies assessing results16,22 although, information on cumulative UV
photocarcinogenesis in psoriasis patients receiving doses received by patients have not been docu-
NB-UVB did not find an increased risk of melanoma mented in these studies.
or KC.8 This review included Hearn et al who studied In contrast, PUVA is considered to be carcinogenic.
Scottish patients undergoing NB-UVB (some with a A Swedish study with mean follow-up of 7.2 years
history of previous PUVA) with a median follow-up and a 5.7-year US prospective study both showed
of 5.5 years, and found no increased incidence of first dose-dependent increased risk of SCC with
skin cancers.13 A Korean population-based study on PUVA.42,43 Using regression models, the risk of
vitiligo patients found no increased risk of KC or melanoma was higher after 250 treatments or
melanoma but did find increased risk of actinic 15 years.44
J AM ACAD DERMATOL Wang et al 765
VOLUME 90, NUMBER 4

Limitations 7. R€unger TM. C--[T transition mutations are not solely UVB-
As with other studies, follow-up duration, and signature mutations, because they are also generated by UVA.
J Invest Dermatol. 2008;128(9):2138-2140.
number of patients are likely limitations and longer- 8. Archier E, Devaux S, Castela E, et al. Carcinogenic risks of
term follow-up will more accurately assess risk of psoralen UV-A therapy and narrowband UV-B therapy in
skin cancer in this population. Although a large range chronic plaque psoriasis: a systematic literature review. J Eur
of treatment exposures was received, median num- Acad Dermatol Venereol. 2012;26(Suppl 3):22-31.
ber of treatments was relatively low at 43 (1-3598). 9. Lee E, Koo J, Berger T. UVB phototherapy and skin cancer risk:
a review of the literature. Int J Dermatol. 2005;44(5):355-360.
Advantages of our study include verification of 10. Wang E, Sasaki J, Nakamura M, Koo J. Cutaneous carcinogenic
skin cancers with pathology reports and evaluation risk of phototherapy: an updated comprehensive review. J
of multiple skin cancers as compared to studies Psoriasis Psoriatic Arthritis. 2015;1:44-51.
which have only looked at first-registered skin 11. Maughan WZ, Muller SA, Perry HO, Pittelkow MR, O’Brien PC.
cancers. This is also one of very few studies Incidence of skin cancers in patients with atopic dermatitis
treated with coal tar. A 25-year follow-up study. J Am Acad
evaluating skin cancer risk with different modalities Dermatol. 1980;3(6):612-615.
of phototherapy (BB-UVB, NB-UVB, and UVAB); 12. Black RJ, Gavin AT. Photocarcinogenic risk of narrowband
with cumulative UV dose through obtaining treat- ultraviolet B (TL-01) phototherapy: early follow-up data. Br J
ment data for all patients; detailed profiling of skin Dermatol. 2006;154(3):566-567.
cancers with anatomical site, previous skin cancer 13. Hearn RM, Kerr AC, Rahim KF, Ferguson J, Dawe RS. Incidence
of skin cancers in 3867 patients treated with narrow-band
history, and demographic factors; and evaluation of ultraviolet B phototherapy. Br J Dermatol. 2008;159(4):931-935.
confounding factors, for example immunosuppres- 14. Jo SJ, Kwon HH, Choi MR, Youn JI. No evidence for increased
sive medications. Our cohort also includes patients skin cancer risk in Koreans with skin phototypes III-V treated
with multiple dermatoses and variety of skin photo- with narrowband UVB phototherapy. Acta Derm Venereol.
types, in contrast to existing studies evaluating 2011;91(1):40-43.
15. Ortiz-Salvador JM, Ferrer DS, Saneleuterio-Temporal M, et al.
Taiwanese and Korean populations who are likely Photocarcinogenic risk associated with narrowband UV-B
to have darker skin phenotypes. phototherapy: an epidemiologic study in a tertiary care
hospital. Actas Dermosifiliogr (Engl Ed). 2018;109(4):340-345.
16. Bae JM, Ju HJ, Lee RW, et al. Evaluation for skin cancer and
CONCLUSION precancer in patients with vitiligo treated with long-term
No increased risk of melanoma, SCC, and BCC narrowband UV-B phototherapy. JAMA Dermatology. 2020;
compared to the general population was found in 156(5):529-537.
17. Man I, Crombie IK, Dawe RS, Ibbotson SH, Ferguson J. The
patients receiving phototherapy with NB-UVB, BB- photocarcinogenic risk of narrowband UVB (TL-01) photo-
UVB, and UVAB. In addition, no correlation between therapy: early follow-up data. Br J Dermatol. Apr 2005;152(4):
skin cancer risk and total number of treatment 755-757.
sessions or cumulative UV dosage was found. This 18. Osmancevic A, Gillstedt M, Wennberg AM, Larko O. The risk of
suggests ultraviolet phototherapy can be considered skin cancer in psoriasis patients treated with UVB therapy.
Acta Derm Venereol. 2014;94(4):425-430.
a relatively safe treatment option. 19. Maiorino A, De Simone C, Perino F, Caldarola G, Peris K.
Melanoma and non-melanoma skin cancer in psoriatic pa-
tients treated with high-dose phototherapy. J Dermatol Treat.
Conflicts of interest 2016;27(5):443-447.
None disclosed. 20. Bajdik CD, Gallagher RP, Astrakianakis G, Hill GB, Fincham S,
McLean DI. Non-solar ultraviolet radiation and the risk of basal
REFERENCES and squamous cell skin cancer. British J Cancer. 1996;73(12):
1. Goeckerman W. The treatment of psoriasis. Northwest Med. 1612-1614.
1925;24:229-231. 21. Raone B, Patrizi A, Gurioli C, Gazzola A, Ravaioli GM. Cutaneous
2. Nexmand PH. Clinical Studies’ of Besnier’s Prurigo, Disserta- carcinogenic risk evaluation in 375 patients treated with
tion. Rosenkilde & Bagger; 1948. narrowband-UVB phototherapy: a 15-year experience from
3. Parrish JA, Jaenicke KF. Action spectrum for phototherapy of our institute. Photodermatol Photoimmunol Photomed. 2018;
psoriasis. J Invest Dermatol. 1981;76(5):359-362. 34(5):302-306.
4. Lim HW, Silpa-archa N, Amadi U, Menter A, Van Voorhees AS, 22. Lin TL, Wu CY, Chang YT, et al. Risk of skin cancer in psoriasis
Lebwohl M. Phototherapy in dermatology: a call for action. J patients receiving long-term narrowband ultraviolet photo-
Am Acad Dermatol. 2015;72(6):1078-1080. therapy: results from a Taiwanese population-based cohort
5. Meduri NB, Vandergriff T, Rasmussen H, Jacobe H. Photo- study. Photodermatol Photoimmunol Photomed. 2019;35(3):
therapy in the management of atopic dermatitis: a systematic 164-171.
review. Photodermatol Photoimmunol Photomed. 2007;23(4): 23. Garritsen FM, Brouwer MW, Limpens J, Spuls PI. Photo(chemo)
106-112. therapy in the management of atopic dermatitis: an updated
6. Sidbury R, Davis DM, Cohen DE, et al. Guidelines of care for the systematic review with implications for practice and research.
management of atopic dermatitis: section 3. Management and Br J Dermatol. 2014;170(3):501-513.
treatment with phototherapy and systemic agents. J Am Acad 24. Statistics Canada. Data tables, 1991 Census, 2019. Accessed
Dermatol. 2014;71(2):327-349. February 17, 2020. https://www12.statcan.gc.ca/English/cen
766 Wang et al J AM ACAD DERMATOL
APRIL 2024

sus91/data/tables/Rp-eng.cfm?LANG=E&APATH=3&DETAIL=1& population and in patients with or without atopic dermatitis

DIM=0&FL=A&FREE=1&GC=0&GID=0&GK=0&GRP=1&PID=32 in the UK. Br J Dermatol. 2010;163(5):1036-1043.
&PRID=0&PTYPE=4&S=0&SHOWALL=No&SUB=0&Temporal= 35. Olesen AB, Engholm G, Storm HH, Thestrup-Pedersen K. The
1991&THEME=101&VID=0&VNAMEE=&VNAMEF= risk of cancer among patients previously hospitalized for
25. McLean DI, Phillips N, Zhou Y, Gallagher R, Lee TK. 40-year atopic dermatitis. J Invest Dermatol. 2005;125(3):445-449.
trends in skin cancer in British Columbia, Canada, 1973 to 36. Hagstr€omer L, Ye W, Nyren O, Emtestam L. Incidence of cancer
2003. J Cutan Med Surg. 2012;16(2):83-91. among patients with atopic dermatitis. Arch Dermatol. 2005;
26. Ahad T, Wang EY, Liu YA, et al. Incidence of skin cancers in 141(9):1123-1127.
patients with eczema treated with ultraviolet phototherapy. J 37. Wang H, Diepgen TL. Atopic dermatitis and cancer risk. Br J
Am Acad Dermatol. 2022;87(2):387-389. Dermatol. 2006;154(2):205-210.
27. Diffey BL, Farr PM. The challenge of follow-up in narrowband 38. Legendre L, Barnetche T, Mazereeuw-Hautier J, Meyer N,
ultraviolet B phototherapy. Br J Dermatol. 2007;157(2):344-349. Murrell D, Paul C. Risk of lymphoma in patients with atopic
28. Chia BKY, Yew YW, Zhao X, Chong WS, Thng TGS. Incidence of dermatitis and the role of topical treatment: a systematic review
skin malignancies in patients with vitiligo or psoriasis who and meta-analysis. J Am Acad Dermatol. 2015;72(6):992-1002.
received narrowband ultraviolet B phototherapy 39. Ming ME, Levy R, Hoffstad O, et al. The lack of a relationship
(308 nm/311 nm): a retrospective review of 3730 patients. between atopic dermatitis and nonmelanoma skin cancers. J
Photodermatol Photoimmunol Photomed. 2022;39:343-350. Am Acad Dermatol. 2004;50(3):357-362.
29. Ko MJ, Huang JW, Wu HY, et al. Risk of skin cancer among 40. Jensen AO, Svaerke C, Farkas DK, Olesen AB, Kragballe K,
patients with chronic kidney disease treated with ultraviolet B Sørensen HT. Atopic dermatitis and risk of skin cancer. Am J
phototherapy for uraemic pruritus: a nationwide cohort Study. Clin Dermatol. 2012;13(1):29-36.
Acta Derm Venereol. 2021;101(2):adv00390. 41. Ling TC, Clayton TH, Crawley J, et al. British Association of
30. Pittelkow MR, Perry HO, Muller SA, Maughan WZ, O’Brien PC. Dermatologists and British Photodermatology Group guide-
Skin cancer in patients with psoriasis treated with coal tar. A lines for the safe and effective use of psoraleneultraviolet A
25-year follow-up study. Arch Dermatol. 1981;117(8):465-468. therapy 2015. Br J Dermatol. 2016;174(1):24-55.
31. Halprin KM, Comerford M, Taylor JR. Cancer in patients with 42. Lindel€of B, Sigurgeirsson B, Tegner E, et al. PUVA and cancer: a
psoriasis. J Am Acad Dermatol. 1982;7(5):633-638. large-scale epidemiological study. Lancet. 1991;338(8759):91-
32. Hannuksela-Svahn A, Pukkala E, Laara E, Poikolainen K, 93.
Karvonen J. Psoriasis, its treatment, and cancer in a cohort 43. Stern RS, Laird N, Melski J, Parrish JA, Fitzpatrick TB, Bleich HL.
of finnish patients. J Invest Dermatol. 2000;114(3):587-590. Cutaneous squamous-cell carcinoma in patients treated with
33. Bhate SM, Sharpe GR, Marks JM, Shuster S, Ross WM. PUVA. N Engl J Med. 1984;310(18):1156-1161.
Prevalence of skin and other cancers in patients with psoriasis. 44. Stern RS, Nichols KT, V€akev€a LH. Malignant melanoma in
Clin Exp Dermatol. 1993;18(5):401-404. patients treated for psoriasis with methoxsalen (psoralen) and
34. Arana A, Wentworth CE, Fernandez-Vidaurre C, Schlienger RG, ultraviolet A radiation (PUVA). N Engl J Med. 1997;336(15):
Conde E, Arellano FM. Incidence of cancer in the general 1041-1045.