Regular PaperJ. Radiat. Res., 52, 496–501 (2011)

Association between Skin Phototype and Radiation Dermatitis in Patients with Breast Cancer Treated with Breast-conserving Therapy: Suntan

Reaction could be a Good Predictor for Radiation Pigmentation

Hideya YAMAZAKI1*, Ken YOSHIDA2, Takuya NISHIMURA1, Kana KOBAYASHI1,Takuji TSUBOKURA1, Naohiro KODANI1, Norihiro AIBE1

and Tsunehiko NISHIMURA1

Breast-conserving therapy/Breast cancer/Radiation dermatitis.The purpose of this study was to evaluate the significance of skin phototype (suntan or sunburn type)

in association with radiation dermatitis in patients with breast cancer who underwent postoperative radio-therapy after breast-conserving surgery because phototype could predict sunlight reaction. We divided patients into two phototypes (58 suntan/darkening and 28 sunburn/reddening types) according to self-reports before radiotherapy. We examined skin color changes in 86 patients who underwent breast-conserving surgery followed by 50 Gy/25 fractions (median) of radiotherapy with or without boost radia-tion (10 Gy/5 fractions). Color change was assessed according to CIE L*a*b* space, which is defined by the Commission Internationale de l’Éclairage (CIE) in 1976 for quantitative color assessment. The patients were also assessed by individual typology angle (ITA°; indicator of skin color calculated by L*a*b* space) and Common Terminology Criteria for Adverse Event v3.0 (CTCAE v3). Radiation therapy changed all values except the b* value, and the suntan type showed a greater darkening response associ-ated with radiation dermatitis than did the sunburn type in terms of ITA° value change (p = 0.04), whereas the sunburn type did not show higher a* value (reddening). By CTCAE v3 classifications, a Grade 2 reac-tion appeared in 14% sunburn patients and in 31% of the suntan group, respectively (p = 0.16). Suntan type predicted higher pigmentation associated with radiation dermatitis. Self-reported phototype has the potential to be a good predictor of skin sensitivity to radiation exposure for clinical screening.

INTRODUCTION

Breast-conserving therapy has become a standard treat-ment as it helps to improve QOL in treating breast cancer patients. Activities of daily living, physical factors, and psy-chological status are the main criteria in assessing QOL,1)

and cosmetic assessment is one of the factors of interest to women. Radiation-induced skin changes were recognized soon after the discovery of x-rays, and were scientifically reported as early as 1896.2) This is a consequence of the well-known fact that sensitivity to radiation varies between

individuals, and underlines the necessity of finding an objec-tive assessment method. Several objective assessment tools have been used for their reliance and reproducibility in der-matology and in industry,3–6) and we have employed chro-mameter to examine radiation dermatitis in an objective manner as a tool for comparison across different radiother-apy schedules.7,8) We used CIE L*a*b* space, which is defined by the Commission Internationale de l’Éclairage (CIE) in 1976 for quantitative color assessment.

Radiation dermatitis is sometimes regarded as a phenom-enon similar to sunlight reaction; however, there is little information to date that compares sunlight reaction and radi-ation dermatitis. Skin phototype classifications were initially developed by Fitzpatrick in an attempt to predict the skin reactions of patients to phototherapy by estimating the skin reaction to sun exposure based on the patient’s reported his-tory, and skin types are categorized as type I (always burns, never tans) to IV (never burns, tans easily).9) These photo-type classifications reflect individual sensitivity and predict the skin reactions of patients to phototherapy, and are now widely used to advise individuals with regard to sun protec-

*Corresponding author: Phone: +81-75-251-5618, Fax: +81-75-251-5840,E-mail: hideya10@hotmail.com

1Department of Radiology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, 465 Kajiicho Kawaramachi Hirokoji, Kamigyo-ku, Kyoto 602-8566 Japan; 2Department of Radiology, National Hospital Organization, Osaka National Hospital, Hoenzaka 2-1-14 Chuo-ku, Osaka city, Osaka 540-0006 Japan.doi:10.1269/jrr.10169

Phototype Analysis for Radiation Pigmentation in BCT 497

tion and to estimate risk category for cutaneous melanoma. Skin color typing was determined by individual typology angle (ITA°) classification (very light to black) calculated by CIE L*a*b* space. In accordance with this notion, we explored the correlation between self-reported sunburn ten-dency or tanning ability and objective color change in radi-ation dermatitis. To simplify the examination, we divided patients into two types according to self-reported sun-reaction (suntan and sunburn).

PATIETNS AND METHODS

Breast irradiation after breast-conserving surgery (BCS) was administered to 316 breast cancer patients at the Osaka National Hospital between May 2000 and June 2005. Of these, 118 patients gave their informed consent for an objec-tive assessment of skin color change. Furthermore, we divided 86 patients into a suntan type (darken skin and last-ing pigmentation) or sunburn type (reddening and rapid recover or slightly burned skin) according to their past light reaction for one hour sun-exposure during summer by self-reported declaration. Physician checked their declaration by visual inspection on breast skin and mole or freckle. All of

the patients were women between the ages of 25 and 78 years (median age: 49 years). Patient characteristics are shown in Table 1.

Of the patients who underwent BCS, the remaining breast was conventionally treated (1.8–2 Gy/day, 5 fractions/week) with a tangential-field 4-MV photon beam for a total dosage of 48.4 50 Gy/25–28 fractions at the reference point. An additional fractionated boost radiation of 10 Gy using 4–10 MeV electrons was administered to patients with a positive surgical margin of 5 mm or less. Wedge filters were used to obtain homogeneous dose distribution. The median follow-up period was 27 months (13–55 months).

We measured the breast skin at room temperature and under room light with a Color Reader CR-13 (Konica-Minolta, Tokyo, Japan). The first measurements were performed before radiotherapy, and subsequent mea-surements just after completion of radiotherapy, then one month, six months and one year after radiotherapy. Color was determined in terms of L*a*b* values, within the three-dimensional CIE L*a*b* space defined by the Commission Internationale de l’Éclairage (CIE) in 1976. Skin color typ-ing or grading of the ITA° for each subject was calculated based on the measured L* and b* values according to the

Table 1. Patients Characteristics

Strata Variables (%) p-value

Age 49 ± 11

BMI 22.5 ± 3.4

T category cTis 11 (13%)

cT1 45 (52%)

cT2 30 (35%)

N category N0 51 (59%)

N1 27 (31%)

N2 8 ( 9%)

Boost / Prescribed dose without boost / 48.4–50 Gy 76 (88%)

with boost / 59.5–60 Gy 10 (12%)

Irradiated field X-axis (cm) 18.2 ± 0.8

Y-axis (cm) 6.9 ± 0.7

Depth (cm) 5.7 ± 0.6

CTX – 50 (58%)

+ 36 (42%)

ITA° Skin color category Treated side Contra-lateral side

(before RT) I 2 ( 2%) 19 (22%)

II 32 (37%) 47 (55%)

III 45 (52%) 20 (23%)

IV 2 ( 2%)

V 5 ( 6%) p < 0.01

ITA°, individual typology angle; BMI, Body weight (kg)/height (m)2

H. Yamazaki et al.498

following formula: ITA° = [ArcTan((L*-50)/b*)]×180/3.14159. The higher the ITA° values, the lighter the skin. Skin color was graded based on the calculated ITA° values and the skin color categories described by Chardon et al. (Table 2).4)

The vertical L* axis (from 0 black to 100 white) repre-sents the luminance of the sample as perceived by the human

eye, the b* axis represents the yellow (> 0)/blue (< 0) com-plementary components of the color, and the a* axis the complementary red (> 0)/green (< 0) components. In terms of skin color, erythema will make skin darker and redder, resulting in a reduction in the L* value and an increase in the a* value. When the a* and b* values are nil, the color is a gray scale gradation. We investigated the respective time courses for color change. Simultaneously, we investigated any correlations of the objective measurements with the Common Terminology Criteria for Adverse Event (CTCAE) v3.0, which has become the standard for evaluating radio-therapy skin reactions. The independent observer who assessed skin reactions using the CTCAE v3 was blind to the results obtained with the colorimeter. All patients were enrolled in the study after their informed consent had been obtained prior to radiotherapy according to the guidelines of the intramural ethics committee.

Statistical AnalysisAll statistical analyses were carried out with the Statview-

v5.0 software program. Student’s t-test was used for normal-ly distributed data and the Mann Whitney U-test for skewed data. Percentages were analyzed with the Chi-square test. For multiple longitudinal data sets, analysis of variance

Table 2. Gradation of skin color recommended by Chardon Grading

ITA° Skin color category

Grade Definition Color type

I > 55° Very light

II > 41 and ≤ 55° Light

III > 28 and ≤ 41° Intermediate

IV > 10 and ≤ 28° Tan/malt

V > –30 and ≤ 10° Brown

VI ≤ 30° Black

ITA°; individual typology angle

Table 3. Patients background between suntan and sunburn group

Strata VariablesSunburn(n = 28)

(%)Suntan(n = 58)

(%) p-value

Age 52 ± 10 48 ± 11 ns (0.08)

BMI 23.1 ± 4.4 22.2 ± 2.8 ns

T category cTis 2 ( 7%) 9 (16%)

cT1 14 (50%) 31 (53%)

cT2 12 (43%) 18 (31%) ns

N category N0 13 (46%) 38 (66%)

N1 10 (36%) 17 (29%)

N2 5 (18%) 3 ( 5%) ns

Boost / Prescribed dose without boost / 48.4–50 Gy 25 (89%) 51 (88%)

with boost / 59.5–60 Gy 3 (11%) 7 (12%) ns

Irradiated field X-axis (cm) 18.1 ± 0.7 18.3 ± 0.7 ns

Y-axis (cm) 6.8 ± 0.8 6.9 ± 0.8 ns

Depth (cm) 5.6 ± 0.7 5.7 ± 0.6 ns

CTX – 13 (46%) 37 (64%)

+ 15 (54%) 21 (36%) ns

ITA° Skin color category Treated side Contra-lateral side Treated side Contra-lateral side

(before RT) I 1 ( 4%) 10 (36%) 1 ( 2%) 9 (16%)Treated side ns

II 13 (46%) 16 (57%) 19 (33%) 31 (53%)

III 13 (46%) 2 ( 7%) 32 (55%) 18 (31%)

IV 1 ( 4%) 1 ( 2%) Contra-lateralside p = 0.02

V 5 ( 9%)

CTCAE v3.0, Common Terminology Criteria for Adverse Event version 3 score; CTX, chemotherapy; ITA°, individual typology angle; BMI, Body weight (kg)/height (m)2

Phototype Analysis for Radiation Pigmentation in BCT 499

(ANOVA) with repeated measurements was also applied. A value of p < 0.05 was considered to be statistically sig-nificant.

RESULTS

Skin color was examined in terms of mean L*a*b* values and ITA°. For ITA° classification, most of the Japanese patients had a “light” or “intermediate” skin color before radiotherapy, namely grade II (37%) and grade III (52%); the others (grades I, IV, V: very light, tan/malt, brown, resp.) accounted for 2–6% (Table 1). In addition, contra-lateral intact breast skin color was found to be “light” grade II (57%), “very light” grade I (22%) and ”intermediate” grade III (23%)(Table 1). Surgical influences remained even before postoperative radiotherapy (p < 0.01), though they were difficult to assess by visual inspection.

For ITA° classification, skin color of treated breast were categorized as 1 grade I (4%), 13 grade II (46%), 13 grade

III (46%), and 1 grade IV (4%) patients in sunburn group, whereas 1 (2%), 19 (33%), 32 (55%) , 1 (2%), and 5 (9%) patients in suntan group (Table 3, n.s.). On the other hands, contra-lateral intact breast skin color was distributed 10 grade I (36%), 16 grade II (57%), 2 grade III (7%) in sun-burn group, which is lighter than the population 9 grade I (16%), 31 grade II (43%), 18 grade III (31%) in suntan group (p = 0.02, Table 3). Other than skin color, there is no difference in background factor distribution between these two groups, including body mass index and treatment fac-tors.

Radiotherapy induced significant alterations in L*, a*, ITA° values (Fig. 1), but not in the b* value. Suntan group showed larger alteration in ITA° value (p < 0.05), not in other values (L*, a*, b* values). Therefore, suntan type patients showed a greater darkening response associated with radiation dermatitis than did the sunburn type (Fig. 1a), whereas the sunburn type did not show a greater reddening response to radiation dermatitis than did the sunburn type

(a)

(b)

(c)

(d)

Fig. 1. Longitudinal alterations for color values over time between suntan and sunburn types. Each symbol represents average value, with error bars showing one SD. As all values represent significant differences between treated and control side, surgery left significant influence on skin color even before radiotherapy. All analyzed factors in treated skin changed in statistically significant manners during the time course by external radiotherapy excluding b* value (L* darker, a* reddish, p < 0.0001; ANOVA). In comparison between suntan type and sunburn type, ITA° value showed significant difference (p = 0.04). Each p-value or NS in figures depicts statistical value by ANOVA between suntan and sunburn group. Abbreviations; Control; contra-lateral untreated breast *represents statistical significance p < 0.05 in comparison between two points. a) ITA° value, b) L* value, c) a* value, d) b* values.

H. Yamazaki et al.500

(Fig. 1c). To assess the role of skin color type on radiation dermatitis, we analyzed ITA° value alteration according to skin type between Type I–II and III–V patients. We define the delta ITA° value = ITA° value (estimated) - ITA° value (before), and found a similar degree of change between lighter and darker constitutive skin groups (Fig. 2).

According to CTCAE v3 toxicity criteria, suntan type showed 31% (18/58) grade 2 reaction, compared to 14% (4/28) in sunburn group (p = 0.16). There are significant differ-ences between grade 1 and grade 2 in L* value (59.3 ± 4.2 vs. 54.3 ± 5.9, p < 0.01), ITA° value (27.3 ± 12.6 vs. 12.6 ±18.8, p < 0.01) and a* value (10.3 ± 1.9 vs. 12.2 ± 2.7, p < 0.01).

DISCUSSION

In radiation dermatitis, generalized erythema, sometimes undetectable without special instruments, may occur hours after radiation exposure, and fade over a period of hours to days.10) A second phase of more sustained erythema is apparent 10 to 14 days after dosing, and is characterized by a blanchable reactive pink hue, without other epidermal changes and most likely mediated by cytokines. Grade 1 changes included follicular or mild generalized erythema and dry desquamation enhanced in the inframammary fold and axilla. Other changes include pruritus, epilation, scaling, and dyspigmentation. The dryness and hair loss are sec-ondary to injury to sebaceous glands and hair follicles. Grade 2 changes, consisting of persistent tender or edema-tous erythema, may progress to focal loss of the epidermis, producing moist desquamation in skin folds. This usually occurs after 4 to 5 weeks with radiation doses to the skin of 40 Gy or more. The time lag occurs during epidermal cells containing melanin moved from the basal layer to stratum corneum.

Sun light reaction was investigated deeply mainly in der-matology. There are different underlying mechanisms in nature for sunburn (reddish color from blood flow elevation) vs. suntan pigmentation (melanin deposition). Sunburn is produced mainly by ultraviolet-B (UVB) over a half-day to two-day period as erythema, telangiectasia, and simulta-neous inflammation with or without fever, buries, and pain. An effusion theory model was hypothesized in which medi-ator substances formed in the epidermis diffuse to the dermis and cause vasodilatation in dermal blood vessels. Suntan occurs over several days to several weeks after UVB radia-tion by activating melanocytes to increase melanin produc-tion. Repeated UVB radiation increases the number of melanocytes, which may also contribute to increased melanin production. Van der Leun et al. reported that the erythema induced by UV radiation with a wavelength of around 300 nm was due to the actions of a diffusion medi-ator arising in the epidermis, whereas radiation at shorter wavelengths, around 250 nm, exerted a direct effect on the dermal blood vessels; this points to a complicated underly-ing mechanism that depends on wavelength.11)

We tried to explore the correlation between sun-light reac-tion and dermatitis. As Fitzpatrick already established phototypes classifications in an attempt to predict the skin reactions of patients to phototherapy by estimating the skin reaction to sun exposure based on the patient’s reported history,9) we applied those methodology. As the skin typing scale was made based on data for Caucasian people, Kawada et al. have proposed Japanese skin types (Japanese Skin Type Classification: J1; easily burns, slight delayed tanning, J2; normal reaction = mildly burn and tan, J3; slight burn and easily tan).12) Accordingly, we divided patients into two categories to simplify the investigation.

Our data indicated that a suntan reaction due to UV expo-sure may have a similar mechanism to radiation dermatitis in terms of pigmentation and reproducible character, which imply that the endogenous radiation sensitivity may be man-ifested in pigmentation. We hypothesize that self-reporting of suntan pigmentation tendency could provide a good indicator for radiation pigmentation. On the other hand, unexpectedly, a reddening reaction did not correlate with previous reaction to sunlight. It is not clear whether the sun-burn reaction has a different mechanism than radiation der-matitis, or if it is difficult to assess the former accurately by subjective questioning. This may also be partly because Japanese women have a relatively smaller breast size and naturally yellowish-light skin, so that any differences due to a mild reaction might be too subtle to observe. In addition, although ITA° value could indicate difference between suntan and sunburn group, L* value did not. It would imply the superiority of ITA ° value including L* and b* values than single L* value in skin color analysis.

Several limitations of phototype estimation should be con-sidered. At first, as it solely depends on subjective manner,

Fig. 2. Delta ITA° value according to constitutive skin color types. Delta ITA° value = ITA° value (estimated) - ITA° value (before).

Phototype Analysis for Radiation Pigmentation in BCT 501

there are several controversial. Several authors reported inconsistency between phototype and sunburn inclination.13–15)

Chan et al. reported that objective measures of pigmentation fail to correlate well with race, whereas race correlates mod-erately with physician-diagnosed skin phototype.14) On the other hands, they related that tanning ability showed a better correlation with skin complexion characteristics than the burning tendency. Next, as CTCAE estimates erythema without consideration to pigmentation, pigmentation dose not influence on CTCAE score. However, we found there are significant differences between Grade 1 and 2 patients in ITA°, and L* value not only in a* value. Therefore, pigmen-tation may have a correlation to reddish change to some extent. These preliminary data should be confirmed by data on a larger number of patients, and a detailed exploration of the mechanisms will be left for further study.

In conclusion, the “suntan” skin phototype predicted higher pigmentation in radiation dermatitis. Self-reported skin phototype has potential as a good predictor of skin sen-sitivity to radiation exposure for clinical screening.

REFERENCES

1. Engel J, et al (2004) Quality of life following breast-conserv-ing therapy or mastectomy: result of a 5-year prospective study. Breast J 10: 223–231.

2. Cipollaro AC and Crossland PM (1967) X-rays and radium in the treatment of diseases of the skin. pp. 15–19, Lea and Febiger; Philadelphia.

3. Damian DL, Halliday GM and Barnetson RS (1997) Predic-tion of minimal erythema dose with a reflectance melanin meter. Br J Dermatol 136: 714–718.

4. Wei L, et al (2007) Skin color measurement in Chinese female population: analysis of 407 cases from 4 major cities of China. Int J Dermatol 46: 835–839.

5. Westerhof W, et al (1990) The relation between constitutional skin color and photosensitivity estimated from UV-induced erythema and pigmentation dose-response curves. J Invest Dermatol 94: 812–816.

6. Guinot C, et al (2005) Sun-reactive skin type in 4912 French adults participating in the SU.VI.MAX study. Photochem Photobiol 81: 934–940.

7. Tanaka E, et al (in press) Objective and longitudinal assess-ment of dermatitis following post-operative Accelerated Partial Breast Irradiation (APBI) using High Dose Rate Inter-stitial Brachytherapy (HDR-ISBT) in patients with breast cancer treated with Breast Conserving Therapy : Reduction of moisture deterioration by APBI. Int J Radiat Oncol Biol Phys.

8. Yamazaki H, et al (2009) Longitudinal practical measurement of skin color and moisture during and after breast conserving therapy: influence of neoadjuvant systemic therapy. Jpn J Radiol 27: 309–315.

9. Fitzpatrick TB (1988) The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 124: 869–871.

10. Hymes SR, Strom EA and Fife C (2006) Radiation dermatitis: clinical presentation, pathophysiology, and treatment 2006. J Am Acad Dermatol 54: 28–46.

11. van der Leun JC (1996) UV radiation from sunlight: sum-mary, conclusions and recommendations. J Photochem Photo-biol 5: 237–244.

12. Kawada A (1986) UVB-induced erythema, delayed tanning, and UVA-induced immediate tanning in Japanese skin. Pho-todermatol 3: 327–333.

13. Rampen FHJ, et al (1988) Unreliability of self-reported burn-ing tendency and tanning ability. Arch Dermatol 124: 885–888.

14. Chan JL, et al (2005) Assessing the role of race in quantitative measures of skin pigmentation and clinical assessments of photosensitivity. J Am Acad Dermatol 52: 609–615.

15. Westerhof W, et al (1990) The relation between constitutional skin color and photosensitivity estimated from UV-induced erythema and pigmentation dose-response curves. J Invest Dermatol 94: 812–816.

Received on November 26, 2010Revision received on February 3, 2011

Accepted on February 12, 2011J-STAGE Advance Publication Date: June 2, 2011