Abstract
Background: Some results from laboratory and epidemiologic studies suggest that diet may influence the risk of melanoma, but convincing evidence for a role of single nutrients or food items is lacking. Diet quality, which considers the combined effect of multiple food items, may be superior for examining this relation.
Objective: We sought to assess whether diet quality, evaluated with the use of 4 different dietary indexes, is associated with melanoma risk.
Methods: In this population-based case-control study, we analyzed the relation between 4 diet quality indexes, the Healthy Eating Index 2010 (HEI-2010), Dietary Approaches to Stop Hypertension (DASH) index, Greek Mediterranean Index (GMI), and Italian Mediterranean Index (IMI), and melanoma risk in a northern Italian community, with the use of data from 380 cases and 719 matched controls who completed a semiquantitative food frequency questionnaire.
Results: In the overall sample, we found an inverse association between disease risk and the HEI-2010 and DASH index, but not the Mediterranean indexes, adjusting for potential confounders (skin phototype, body mass index, energy intake, sunburn history, skin sun reaction, and education). However, in sex stratified analyses, the association appeared only in women (P-trend: 0.10 and 0.04 for the HEI-2010 and DASH index, respectively). The inverse relations were stronger in women younger than age 50 y than in older women, for whom the GMI and IMI scores also showed an inverse association with disease risk (P-trend: 0.05 and 0.02, respectively).
Conclusions: These results suggest that diet quality may play a role in cutaneous melanoma etiology among women.
Keywords: diet, diet quality, dietary patterns, melanoma, case-control study, epidemiology
Introduction
Cutaneous melanoma accounts for ∼4% of skin cancers in the United States but is responsible for nearly 80% of skin cancer deaths and 1–2% of all cancer deaths (1). Its incidence has increased in recent years and this upward trend is expected to continue in many countries, despite primary prevention efforts (2–5). It has been postulated that the trend is due to increased exposure to environmental factors acting together with increased genetic susceptibility (6). Among environmental and lifestyle factors such as sun exposure, diet has recently received considerable attention (7–9). The biological plausibility of an association between diet and melanoma etiology is supported in laboratory models by the antioxidant activity of several dietary factors; considerations include the potential involvement of reactive oxygen species and free radicals in skin carcinogenesis, and the ability of some nutrients to inhibit melanoma cell proliferation (7, 8).
Epidemiologic studies examining the possible relation between dietary habits and risk of cutaneous melanoma, however, have been few and have not yielded consistent results. Recent studies variously report an inverse association of melanoma risk with intake of vitamins C, D, and E, soluble carbohydrates, alcohol, and specific foods and beverages such as fish, vegetables, fruit, tea, and coffee (9–16). Conversely, direct associations have been found with PUFAs, linoleic acid (18:2n–6), and selenium (10–12, 14, 15, 17–20). These inconsistencies may be due to a lack of true association between diet and melanoma risk, biases such as inadequate dietary assessment or uncontrolled confounding, or the methodologic inadequacy of an approach based on single nutrients or foods. Concerning the latter hypothesis, it is possible that an approach based on dietary pattern, which considers the cumulative effects and interactions of multiple food items, may be superior for identifying a relation between diet and melanoma risk (21). Diet quality represents a departure from the traditional focus on single nutrients, providing a better understanding of disease etiology and prevention in nutritional epidemiology, and this explains its growing use in epidemiologic studies (22–28). Two alternative approaches are commonly used in epidemiologic research when dealing with dietary patterns and health risks: an a priori one, based on already-established patterns that have been demonstrated to correlate with the risk of several chronic diseases (including cancer), and an “exploratory,” a posteriori approach newly defined on the basis of the specific data collected within a study. Both approaches have advantages and limitations. Here, we take the a priori approach, which some authors consider a stronger approach (29).
To our knowledge, to date, no epidemiologic studies have assessed the association of a priori–defined dietary patterns on melanoma risk despite the growing awareness of a potential role of diet in its etiology, and only one randomized trial has addressed the effect of complex dietary changes on melanoma risk (30). In that trial, carried out in the United States, a low-fat diet was associated with increased melanoma risk among those with higher baseline fat intake and decreased risk among those with lower baseline fat intake (30). In the present population-based case-control study, we investigated the risk of cutaneous melanoma in association with 3 well-established a priori diet-quality indexes, the Healthy Eating Index 2010 (HEI-2010)13 (31), the Dietary Approaches to Stop Hypertension (DASH) diet (32), and the Greek Mediterranean Index (GMI) (33), and with another pattern we recently defined, the Italian Mediterranean Index (IMI) (34, 35). All of these dietary patterns have been related to a decreased risk of several chronic diseases, such as cardiovascular disease and cancer (36–38). The potential etiologic role of a Mediterranean diet is also of special interest for melanoma, considering that its incidence in the southern Italian population is about one-half that observed in northern Italy (39), and that Greece and Albania show the lowest melanoma incidence in Europe (40).
Methods
Details of our case-control study on dietary risk factors of cutaneous melanoma in the population (∼3 million people) of 5 provinces of the Emilia Romagna region in northern Italy have been provided elsewhere (13, 15). Briefly, in the years 2005–2006, we attempted to recruit all patients with newly diagnosed cutaneous malignant melanoma residing in the provinces of Bologna, Ferrara, Modena, Parma, and Reggio Emilia and being seen at local dermatology clinics. Inclusion criteria were a histologically based diagnosis of cutaneous malignant melanoma without clinical evidence of metastasis. Five hundred seventy-two eligible patients were contacted by their dermatologists to participate in the study immediately after diagnosis, and 394 (69%) agreed to participate. Each subject was administered lifestyle questionnaires and FFQs (see below) when attending a routine visit at the beginning of follow-up care for their disease, in all cases within 1 mo of the date of melanoma excision and usually at the time of communication of the diagnosis. The FFQ questionnaire could be returned at the following visit or sent back to the Center for Environmental, Genetic, and Nutritional Epidemiology at Modena and Reggio Emilia University. Six referents matched to each case for gender, year of birth (±5 y), and province of residence were randomly selected from the database of the Emilia-Romagna region National Health Service directory (mandatory for all Italian residents) after enrollment of cases. An envelope containing a lifestyle questionnaire and FFQ, a description of the study, and a prepaid return envelope were mailed to these 2825 potential controls. Family doctors of these potential controls, also retrieved through the Emilia-Romagna region National Health Service directory, were informed by mail about the study and its purpose to encourage involvement of the potential participants. Seven hundred forty-seven (26%) potential controls agreed to participate in the study and returned the questionnaires. Fourteen cases and 28 controls were excluded from subsequent analysis because of incompleteness of data or extreme values (ratio of total energy intake to calculated basal metabolic rate lower than the 0.5th percentile or higher than the 99.5th percentile) derived from the FFQ. All participants gave their written informed consent before enrollment.
Dietary assessment.
Habitual diet during the previous 12 mo was investigated with the use of a validated semiquantitative FFQ designed to capture eating behaviors in Italy, specifically developed as part of the EPIC (European Prospective Investigation into Cancer and Nutrition) study for the northern Italian population (41–43). The EPIC questionnaire was designed to be self-administered and was checked by trained personnel after compilation. Participants were asked to respond to 248 questions about 188 different food items, including seasonal foodstuffs, and to indicate the number of times a given item was consumed (per day, week, month, or year), from which the absolute frequency of consumption of each item was calculated.
The quantity of food consumed was assessed by selecting an image of a food portion or selection of a predefined standard portion when no image was available. The food items were then linked with the use of specially designed software (42, 43) to the Italian Food Tables (44) to obtain estimates of daily intake of macro- and micronutrients plus energy.
Diet quality indexes.
For each study participant, we computed scores for 4 a priori–defined diet quality indexes. Components and standard portions for the scoring of each index are provided in Supplemental Table 1.
Two of the diet quality indexes were originally developed in the United States and have been shown to be inversely associated with cancer risk on the basis of sound epidemiologic evidence (36, 38, 45). The HEI-2010 was originally designed to monitor the diet quality of the US population and compliance with the 2010 Dietary Guidelines for Americans and can be applied to any defined set of foods and menu (31, 46). It has been used for epidemiologic research to examine relations between diet and health-related outcomes and the effectiveness of nutrition intervention programs (46). We adjusted the food quantities in our database to the HEI-2010 quantities with the use of the MyPyramid Equivalents Database, Version 2.0 (USDA). The HEI-2010 has 12 components, each assessed as servings per 4.18 MJ (1000 kilocalories): total fruit, whole fruit, total vegetables, dark green vegetables and beans, whole grains, refined grains, dairy, total protein foods, seafood and plant proteins, FAs (ratio of PUFAs and MUFAs to SFAs), sodium, and empty calories (calories from solid fats, alcohol, and added sugars). The intake of components is scored such that a maximum score indicates optimal consumption. For most components, a higher score indicates higher consumption, whereas for refined grains, sodium, and empty calories, a higher score indicates lower consumption. For total fruit, whole fruit, total vegetables, dark green vegetables and beans, total protein foods, and seafood and plant proteins, the scores were 0–5. For whole grains, dairy, refined grains, and sodium, the scores were 0–10. For empty calories, the score was 0–20. The range of possible total scores was 0–100.
The DASH diet was originally designed to reduce blood pressure and has been demonstrated to be effective in this regard (32, 47). It is considered to be an effective healthy eating pattern according to several studies (36, 45, 48, 49). We scored adherence to the DASH diet as suggested by Fung et al. (50). The score is based on 8 components: fruits, vegetables, nuts and legumes, low-fat dairy products, and whole grains, with high intake yielding high DASH scores, and sodium, sweetened beverages, and red and processed meats, with low intake yielding high scores. Participants were classified into quintiles according to intake of each component and the component scores were summed to obtain the DASH score (overall possible range 8–40).
The third and fourth patterns, the GMI and IMI, were formulated as Mediterranean diet scales (33, 35), taking into account the large body of evidence supporting the beneficial effect of a Mediterranean diet in preventing cancer and other chronic disease (37, 51). For the GMI, scoring is based on intake of 9 items: vegetables, legumes, fruit and nuts, dairy products, cereals, meat and meat products, fish, alcohol, and the ratio of MUFAs to SFAs. For most items, consumption above the study median received 1 point; all other intake received 0 points. For dairy products and meat and meat products, consumption below the median received 1 point. Medians were gender specific. For ethanol, men who consumed 10–50 g/d and women who consumed 5–25 g/d received 1 point; otherwise, the score was 0. The range of possible scores was 0–9.
The IMI pattern was recently developed by the Epidemiology and Prevention Unit of the Milan National Cancer Institute (35) by adapting the GMI to typical Italian eating behavior. The score is based on intake of 11 items: 6 typical Mediterranean foods or food groups (pasta; typical Mediterranean vegetables such as raw tomatoes, leafy vegetables, onion and garlic, salad, and fruiting vegetables; fruit; legumes; olive oil; and fish), 4 non-Mediterranean foods (soft drinks, butter, red meat, and potatoes), and alcohol. One point is received for consumption of each typical Mediterranean food in the upper tertile of the distribution, and for consumption of each non-Mediterranean food in the bottom tertile; all other intake received 0 points. Ethanol received 1 point for intake up to 12 g/d; abstainers and persons who consumed >12 g/d received a 0 score. The range of possible scores was 0–11.
Additional variables.
In addition to the EPIC FFQ, each participant completed a questionnaire collecting additional information. The questionnaire included items on demographic characteristics (place and date of birth, province of residence, marital status, and education), self-reported weight and height, phenotypic characteristics (eye, hair, and skin color), sunburn history (never, only before, or only after 18 y of age) and skin sun reaction (speed of tan and tendency to burn). Eye color was classified into light (blue/green), light brown, and dark (brown/black) categories. Hair color was classified as blond, red, light brown, or dark brown/black at 20 y, and skin color was classified as white, light brown, brown/olive, or dark brown/ebony. Each subject was assigned to the corresponding skin phototype with the use of the Fitzpatrick phototyping scale (52).
Statistical analysis.
All melanoma cases were matched to one or more controls on province of residence, sex, and age (± 5 y). We compared the frequency of categorical variables between cases and controls with the use of chi-square tests, and we assessed the correlations among the 4 diet quality indexes with the use of Pearson correlation coefficients. We defined exposure categories by setting a priori cutoffs of diet quality scores (53). We used conditional logistic regression to obtain ORs and 95% CIs of cutaneous melanoma associated with diet quality score, both crude and with adjustment for BMI (continuous), education (4 categories), nonalcoholic energy intake (continuous), skin phototype (4 categories), sunburn history (3 categories) and skin sun reaction (5 categories). Because unadjusted and adjusted ORs were similar, only adjusted results are presented. We also tested for linear trend by entering diet quality scores as continuous variables in conditional logistic regression models (54). In these analyses, a 1-unit increase was defined as a 10-point increase in HEI-2010, a 3-point increase in DASH index, and a 1-point increase in the GMI or IMI; these increments each correspond to ∼10% of the total score range.
We also investigated age, gender, and skin phototype as potential effect modifiers by conducting analyses stratified on these factors (4, 55–57). Because these analyses involved smaller sample sizes, rather than use cutoffs, which requires more df, we tested for linear trend by entering the diet quality scores as continuous variables in conditional logistic regression models.
Results
Three hundred eighty patients, 175 men and 205 women with mean age 58 ± 16 and 53 ± 15, respectively, and 719 age-, sex-, and province of residence–matched referents were included in the analysis. Demographic and clinical characteristics of cases and controls are reported in Table 1. Educational attainment and marital status were similar in the 2 groups (chi-square P = 0.97 and 0.46, respectively). Cases tended to have more fair skin types and were more likely to report a history of sunburns and difficulty tanning (all comparisons, P < 0.01). The overall distribution of diet quality scores in the participants, reported in Table 1, was similar to that observed in other Italian populations (35). The 4 dietary index scores were moderately correlated, with the highest correlation between the HEI-2010 and DASH index (Table 2).
TABLE 1.
Characteristics of study subjects1
| Cases | Controls | |
| Gender | ||
| Men | 175 (46.1) | 319 (44.4) |
| Women | 205 (53.9) | 400 (55.6) |
| Age, y | ||
| <40 | 108 (15.0) | 64 (13.0) |
| 40–59 | 306 (42.5) | 170 (34.4) |
| 60–79 | 278 (38.7) | 235 (47.6) |
| ≥80 | 27 (3.8) | 25 (5.6) |
| Education | ||
| Elementary or less | 91 (23.9) | 170 (23.6) |
| Middle school | 95 (25.0) | 176 (24.5) |
| High school | 136 (35.8) | 266 (37.0) |
| College or more | 54 (15.3) | 107 (14.9) |
| Marital status | ||
| Married | 257 (67.6) | 493 (68.7) |
| Unmarried/single | 68 (17.9) | 103 (14.3) |
| Divorced | 23 (6.0) | 48 (6.6) |
| Widowed | 31 (8.2) | 74 (10.3) |
| Unknown | 1 (0.3) | 1 (0.1) |
| BMI, kg/m2 | ||
| <20 | 31 (8.2) | 45 (6.3) |
| ≥20 to <25 | 164 (43.2) | 306 (42.7) |
| ≥25 to <30 | 133 (35.0) | 287(39.9) |
| ≥30 | 52 (13.6) | 81 (11.3) |
| Phototype | ||
| I | 105 (27.6) | 109 (15.2) |
| II | 136 (35.8) | 238 (33.1) |
| III | 122 (32.1) | 312 (43.4) |
| IV | 17 (4.5) | 60 (8.3) |
| Sunburn history | ||
| Never | 182 (47.9) | 452 (62.9) |
| Before 18 y | 108 (28.4) | 164 (22.8) |
| After 18 y | 90 (23.7) | 103 (14.3) |
| Skin sun reaction (speed to tan and tendency to burn) | ||
| High tendency to burn and never tan | 21 (5.5) | 22 (3.1) |
| High tendency to burn and moderate tan | 117 (30.8) | 131 (18.2) |
| Moderate tendency to burn and gradual tan | 160 (42.1) | 357 (49.7) |
| No tendency to burn and golden tan | 65 (17.1) | 149 (20.7) |
| No tendency to burn and intense tan | 17 (4.5) | 60 (8.3) |
| Dietary pattern score, median (range) | ||
| HEI-2010 | 57 (22–92) | 59 (22–95) |
| DASH | 23 (8–36) | 24 (11–36) |
| GMI | 4 (1–9) | 5 (0–9) |
| IMI | 4 (0–10) | 4 (0–9) |
Values are n (%) unless otherwise indicated; overall n = 380 (34.6%) for cases and n = 719 (65.4%) for controls. DASH, Dietary Approaches to Stop Hypertension; GMI, Greek Mediterranean Index; HEI-2010, Healthy Eating Index 2010; IMI, Italian Mediterranean Index; Phototype I, eyes/hair/skin light, high tendency to burn and never/moderate tan; Phototype II, eyes/hair/skin light, moderate tendency to burn and gradual tan or eyes/hair/skin brown, high tendency to burn and moderate tan; Phototype III, eyes/hair/skin brown, moderate/no tendency to burn and gradual/golden tan; Phototype IV, no tendency to burn and intense tan.
TABLE 2.
Pearson correlation coefficients (95% CIs) among the scores of the dietary indexes in the men and women studied1
| DASH | GMI | IMI | |
| HEI-2010 | 0.73 (0.71, 0.76) | 0.52 (0.47, 0.56) | 0.54 (0.49, 0.58) |
| DASH | 0.53 (0.49, 0.57) | 0.60 (0.56, 0.64) | |
| GMI | 0.54 (0.50, 0.58) |
P < 0.001 for all values. DASH, Dietary Approaches to Stop Hypertension; GMI, Greek Mediterranean Index; HEI-2010, Healthy Eating Index 2010; IMI, Italian Mediterranean Index.
Adjusted ORs for melanoma for each dietary index are shown in Table 3, both in the whole sample and by sex. In the overall sample, ORs for the highest score category compared with the lowest were consistent with decreased risk for all 4 indexes. Analysis for linear trend on continuous values, however, suggested decreasing risk with higher diet quality only for the HEI-2010 and DASH index. In sex-stratified analyses, among men, there was no evidence of decreasing risk with increasing scores, aside from a low OR in the highest category of the DASH index (OR: 0.88; 95% CI: 0.40, 1.93) and IMI (OR: 0.65; 95% CI: 0.28, 1.50). Among women, trend analyses suggested possible inverse associations with disease risk for the HEI-2010 (OR: 0.90; 95 CI%: 0.34, 2.39; P = 0.10) and DASH index (OR: 0.58; 95% CI: 0.30, 1.13; P = 0.04), whereas a consistent association across all categories of exposure emerged only for the DASH diet.
TABLE 3.
Overall and sex-stratified adjusted ORs (95% CIs) for cutaneous melanoma by level of adherence to dietary pattern scores from a case-control study of diet and melanoma in the Emilia-Romagna region, Italy1
| Score categories |
||||||
| Dietary patterns | 1 | 2 | 3 | 4 | OR (95% CI)2 | P-trend3 |
| All subjects (380 cases and 719 controls) | ||||||
| HEI-2010 | 0.03 | |||||
| Score | <40 | 40–59 | 60–79 | ≥80 | ||
| Cases/controls | 37/71 | 178/295 | 145/307 | 20/46 | ||
| OR (95% CI) | 1.00 | 1.19 (0.74, 1.92) | 0.87 (0.54, 1.41) | 0.85 (0.42, 1.74) | 0.91 (0.83, 1.01) | |
| DASH | 0.03 | |||||
| Score | <20 | 20–24 | 25–29 | ≥30 | ||
| Cases/controls | 89/152 | 154/249 | 93/233 | 44/85 | ||
| OR (95% CI) | 1.00 | 1.01 (0.71, 1.45) | 0.68 (0.46, 1.00) | 0.70 (0.42, 1.16) | 0.90 (0.83, 0.98) | |
| GMI | 0.34 | |||||
| Score | 0–2 | 3–4 | 5–6 | 7–9 | ||
| Cases/controls | 61/102 | 140/250 | 128/282 | 51/85 | ||
| OR (95% CI) | 1.00 | 0.82 (0.54, 1.23) | 0.66 (0.43, 1.01) | 0.86 (0.52, 1.45) | 0.96 (0.89, 1.04) | |
| IMI | 0.38 | |||||
| Score | 0–2 | 3–4 | 5–6 | 7–10 | ||
| Cases/controls | 76/159 | 157/263 | 117/222 | 30/75 | ||
| OR (95% CI) | 1.00 | 1.26 (0.88, 1.80) | 1.13 (0.77, 1.65) | 0.76 (0.44, 1.32) | 0.97 (0.90, 1.04) | |
| Men (175 cases and 319 controls) | ||||||
| HEI-2010 | 0.31 | |||||
| Score | <40 | 40–59 | 60–79 | ≥80 | ||
| Cases/controls | 22/40 | 86/142 | 60/126 | 7/11 | ||
| OR (95% CI) | 1.00 | 1.09 (0.57, 2.08) | 0.81 (0.41, 1.59) | 1.26 (0.39, 4.03) | 0.95 (0.82, 1.11) | |
| DASH | 0.42 | |||||
| Score | <20 | 20–24 | 25–29 | ≥30 | ||
| Cases/controls | 42/85 | 79/116 | 35/85 | 19/33 | ||
| OR (95% CI) | 1.00 | 1.38 (0.80, 2.38) | 0.86 (0.46, 1.62) | 0.88 (0.40, 1.93) | 0.95 (0.83, 1.07) | |
| GMI | 0.81 | |||||
| Score | 0–2 | 3–4 | 5–6 | 7–9 | ||
| Cases/controls | 20/42 | 63/110 | 67/119 | 25/48 | ||
| OR (95% CI) | 1.00 | 0.99 (0.49, 1.98) | 0.99 (0.49, 1.98) | 0.93 (0.41, 2.10) | 0.98 (0.87, 1.12) | |
| IMI | 0.43 | |||||
| Score | 0–2 | 3–4 | 5–6 | 7–10 | ||
| Cases/controls | 38/79 | 66/106 | 58/99 | 13/35 | ||
| OR (95% CI) | 1.00 | 1.38 (0.79, 2.40) | 1.16 (0.66, 2.05) | 0.65 (0.28, 1.50) | 0.96 (0.86, 1.07) | |
| Women (205 cases and 400 controls) | ||||||
| HEI-2010 | 0.10 | |||||
| Score | <40 | 40–59 | 60–79 | ≥80 | ||
| Cases/controls | 15/31 | 92/153 | 85/181 | 13/35 | ||
| OR (95% CI) | 1.00 | 1.49 (0.70, 3.17) | 1.08 (0.51, 2.26) | 0.90 (0.34, 2.39) | 0.90 (0.79, 1.02) | |
| DASH | 0.04 | |||||
| Score | <20 | 20–24 | 25–29 | ≥30 | ||
| Cases/controls | 47/67 | 75/133 | 58/148 | 25/52 | ||
| OR (95% CI) | 1.00 | 0.75 (0.46, 1.25) | 0.57 (0.35, 0.95) | 0.58 (0.30, 1.13) | 0.88 (0.79, 0.88) | |
| GMI | 0.23 | |||||
| Score | 0–2 | 3–4 | 5–6 | 7–9 | ||
| Cases/controls | 41/60 | 77/140 | 61/163 | 26/37 | ||
| OR (95% CI) | 1.00 | 0.74 (0.44, 1.25) | 0.49 (0.28, 0.86) | 0.93 (0.46, 1.86) | 0.94 (0.985, 1.05) | |
| IMI | 0.73 | |||||
| Score | 0–2 | 3–4 | 5–6 | 7–10 | ||
| Cases/controls | 38/80 | 91/157 | 59/123 | 17/40 | ||
| OR (95% CI) | 1.00 | 1.25 (0.77, 2.03) | 1.14 (0.67, 1.93) | 0.92 (0.44, 1.93) | 0.98 (0.89, 1.09) | |
ORs adjusted for phototype, BMI, nonalcoholic energy intake, sunburn history, skin sun reaction, and education. DASH, Dietary Approaches to Stop Hypertension; GMI, Greek Mediterranean Index; HEI-2010, Healthy Eating Index 2010; IMI, Italian Mediterranean Index.
Adjusted OR for melanoma for 10-point increase in HEI-2010, 3-point increase in DASH, and 1-point increase GMI and IMI pattern scores.
P value for linear trend with the use of continuous values of dietary index score in conditional logistic regression models.
In age-stratified analysis (cutoff 50 y), there was indication of an inverse association only among the younger subjects (Table 4). When further stratifying for sex, an inverse association was much more evident among younger women than younger men. When stratifying for phototype (I and II vs. III and IV), trend analysis yielded some suggestion of stronger inverse associations between dietary pattern and disease risk among individuals with lighter phototypes compared with darker phototypes (Table 4), although both category-based and continuous ORs had wide CIs.
TABLE 4.
Adjusted ORs (95% CIs) for melanoma according to adherence to dietary pattern in age- and phototype-stratified analysis1
| All cases (380 cases/719 controls) |
Men (175 cases/319 controls) |
Women (205 cases/400 controls) |
||||
| OR (95% CI) | P-trend | OR (95% CI) | P-trend | OR (95% CI) | P-trend | |
| Age stratification | ||||||
| HEI-2010 | ||||||
| <50 y | 0.89 (0.76, 1.04) | 0.08 | 0.90 (0.67, 1.21) | 0.37 | 0.85 (0.70, 1.04) | 0.06 |
| ≥50 y | 0.93 (0.82, 1.06) | 0.22 | 0.97 (0.80, 1.17) | 0.47 | 0.93 (0.78, 1.01) | 0.48 |
| DASH | ||||||
| <50 y | 0.86 (0.76, 0.98) | 0.03 | 0.93 (0.71, 1.20) | 0.74 | 0.80 (0.70, 0.93) | 0.01 |
| ≥50 y | 0.93 (0.83, 1.04) | 0.25 | 0.94 (0.80, 1.10) | 0.36 | 0.94 (0.80, 1.11) | 0.65 |
| GMI | ||||||
| <50 y | 0.87 (0.76, 0.99) | 0.04 | 0.82 (0.62, 1.09) | 0.17 | 0.85 (0.73, 1.00) | 0.05 |
| ≥50 y | 1.03 (0.93, 1.14) | 0.58 | 1.02 (0.88, 1.19) | 0.78 | 1.02 (0.88, 1.18) | 0.84 |
| IMI | ||||||
| <50 y | 0.93 (0.83, 1.06) | 0.27 | 1.04 (0.84, 1.30) | 0.71 | 0.83 (0.70, 0.97) | 0.02 |
| ≥50 y | 0.98 (0.89, 1.08) | 0.74 | 0.90 (0.78, 1.04) | 0.14 | 1.08 (0.94, 1.24) | 0.29 |
| Phototype stratification | ||||||
| HEI-2010 | ||||||
| Phototype I-II | 0.90 (0.78, 1.05) | 0.14 | 1.00 (0.78, 1.28) | 0.78 | 0.86 (0.70, 1.06) | 0.16 |
| Phototype III-IV | 0.90 (0.73, 1.11) | 0.33 | 0.83 (0.57, 1.20) | 0.37 | 0.98 (0.75, 1.28) | 0.74 |
| DASH | ||||||
| Phototype I-II | 0.89 (0.78, 1.01) | 0.09 | 0.89 (0.72, 1.08) | 0.27 | 0.90 (0.76, 1.06) | 0.26 |
| Phototype III-IV | 0.93 (0.79, 1.10) | 0.48 | 1.13 (0.85, 1.49) | 0.50 | 0.84 (0.67, 1.06) | 0.20 |
| GMI | ||||||
| Phototype I-II | 0.96 (0.85, 1.09) | 0.53 | 1.00 (0.82, 1.22) | 0.99 | 0.94 (0.80, 1.11) | 0.47 |
| Phototype III-IV | 0.98 (0.83, 1.15) | 0.77 | 0.97 (0.75, 1.25) | 0.79 | 0.97 (0.78, 1.22) | 0.80 |
| IMI | ||||||
| Phototype I-II | 0.90 (0.79, 1.02) | 0.10 | 1.00 (0.83, 1.20) | 1.00 | 0.82 (0.68, 0.98) | 0.03 |
| Phototype III-IV | 1.03 (0.89, 1.20) | 0.68 | 1.02 (0.80, 1.30) | 0.88 | 1.08 (0.88, 1.34) | 0.44 |
Adjusted for age, phototype, BMI, nonalcoholic energy intake, sunburn history, skin sun reaction, and education. OR for melanoma for 10-point increase in HEI-2010, 3-point increase in DASH, and 1-point increase in GMI and IMI pattern scores. DASH, Dietary Approaches to Stop Hypertension; GMI, Greek Mediterranean Index; HEI-2010, Healthy Eating Index 2010; IMI, Italian Mediterranean Index; Phototype I, eyes/hair/skin light, high tendency to burn and never/moderate tan; Phototype II, eyes/hair/skin light, moderate tendency to burn and gradual tan or eyes/hair/skin brown, high tendency to burn and moderate tan; Phototype III, eyes/hair/skin brown, moderate/no tendency to burn and gradual/golden tan; Phototype IV, no tendency to burn and intense tan.
Discussion
Diet is considered a major risk factor for several cancers, but melanoma generally has not been regarded as a disease influenced by dietary habits. Rather, etiologic importance is attributed to dysplastic nevi, history of sunburns, UV light exposure, phototype, and genetic factors. A few studies have investigated the possible relation between diet and melanoma etiology, most of them in the United States and Italy, and they provided inconsistent evidence of a relation between dietary factors and disease risk (7, 8).
The results of the present study give some support to an inverse association between disease risk and indicators of diet quality, particularly as assessed through the HEI-2010 and DASH index. However, this inverse association was almost entirely due to an association in women, and particularly younger women, suggesting effect modification of gender and age. In fact, all 4 indexes showed an inverse association with disease risk for women under 50 y old. There was a weaker suggestion of effect modification by phototype. The reasons for a gender-specific association between diet quality and melanoma prevention cannot be inferred from the current study. However, our finding of a stronger relation among women younger than 50 y, a premenopausal age group, suggests that hormonal factors could play a role (55, 57). Regardless, our results clearly need to be confirmed by further studies before this conclusion can be reached. Another possible explanation for our findings might be differential diet reporting by sex.
Contrary to our expectations, in our southern European population, the dietary patterns more strongly associated with disease risk in the overall sample were not the scores related to the Mediterranean diet but those assessing diet quality of the US population; that is, the HEI‑2010 and DASH index. The Mediterranean diet indexes have shown associations with risk of other diseases: the Italian index was recently shown to be associated with decreased risk of colorectal cancer and stroke (34, 35), and the Greek index has been associated with decreased mortality from various diseases (58–60). Although the 4 dietary patterns we examined share common characteristics, including an emphasis on high intake of fruit and vegetables, moderate consumption of dairy products, and low intake of animal protein such as red and processed meat, there are important differences. The Mediterranean diet puts emphasis on pasta and fish intake and on moderate consumption of alcohol (with different quantities for the IMI and GMI) (34, 60), whereas the DASH diet and HEI-2010 favor a low consumption of refined grains, sugar-sweetened beverages, and sodium and a high intake of low-fat dairy, with alcohol intake ignored in the DASH diet and penalized in the HEI-2010 (31, 46).
A limitation of our study was the lack of information about type and number of dysplastic nevi or details about occupational history, long-term exposure to UV light, physical activity (a potential marker of vitamin D and UV light exposure), and family history of melanoma; these factors were not assessed because of the risk of recall bias among cases. However, these factors are unlikely to explain the gender-specific associations we detected. Furthermore, we adjusted our ORs for a number of potential confounders, including phototype, BMI, education, nonalcoholic energy intake, history of sunburn, and skin sun reaction, and this adjustment had limited effects on the point estimates, suggesting that major confounding was unlikely. Information about smoking was not collected in the study, but it should not play a major role in melanoma etiology (61–63); therefore, lack of consideration of smoking should not have substantially biased our estimates. Weight and height were self-reported by study participants, which may have introduced some misclassification of BMI.
Diet is not widely regarded as a risk factor for melanoma by the general population, making recall bias in long-term dietary habits unlikely. Moreover, the melanoma cases were recruited at clinical diagnosis of the disease and in the absence of metastases; hence, it is unlikely that the disease itself or its diagnosis induced substantial metabolic abnormalities or change in dietary habits. However, general beliefs about associations between healthy diets and chronic disease may have biased reporting of diet, and thus the occurrence of differential misclassification in women cannot be entirely ruled out. Results from cohort investigations are clearly needed to confirm our findings.
Another limitation of the present study is the dependence on a single assessment—the FFQ—to estimate antecedent dietary habits, with the potential for measurement error because of difficulties in recalling usual dietary habits and insensitivity to long-term dietary changes. However, the FFQ used in this study is a validated tool specifically designed to assess long-term (1 y) usual consumption of as many as 188 different food items. In addition, the sample sizes may have been inadequate to detect some associations.
Although our findings are not sufficiently strong to support recommendations on specific dietary guidelines for melanoma prevention, as has been put forth for other cancers and cardiovascular disease (37, 38, 51, 63), they suggest that an approach focused on indicators of diet quality, rather than on single nutrients or food items, can identify important areas of future research.
Acknowledgments
CM, MM, CMC, CA, SS, VK, GP, and MV designed the study, were responsible for the development of the overall research plan and study oversight, wrote the paper, and had primary responsibility for the final content. CM, CF, FF, CL, CR, GA, AL, LV, AV, CP, MS, PAF, ED, GP, and MV conducted the research by recruiting patients and controls and collecting the data. CM, MM, CA, CMC, SS, VK, and MV performed the statistical analysis. CM, MM, CMC, and MV wrote the paper. All authors read and approved the final manuscript.
Footnotes
Abbreviations used: DASH, Dietary Approaches to Stop Hypertension; EPIC, European Prospective Investigation into Cancer and Nutrition; GMI, Greek Mediterranean Index; HEI-2010, Healthy Eating Index 2010; IMI, Italian Mediterranean Index.
References
- 1.Miller AJ, Mihm MC Jr. Melanoma. N Engl J Med 2006;355:51–65. [DOI] [PubMed] [Google Scholar]
- 2.Pellacani G, Lo Scocco G, Vinceti M, Albertini G, Raccagni AA, Baldassari L, Catrani S, Donelli S, Ghetti P, Lanzoni A, et al. Melanoma epidemic across the millennium: time trends of cutaneous melanoma in Emilia-Romagna (Italy) from 1997 to 2004. J Eur Acad Dermatol Venereol 2008;22:213–8. [DOI] [PubMed] [Google Scholar]
- 3.Wong JR, Harris JK, Rodriguez-Galindo C, Johnson KJ. Incidence of childhood and adolescent melanoma in the United States: 1973–2009. Pediatrics 2013;131:846–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Erdmann F, Lortet-Tieulent J, Schuz J, Zeeb H, Greinert R, Breitbart EW, Bray F. International trends in the incidence of malignant melanoma 1953–2008–are recent generations at higher or lower risk? Int J Cancer 2013;132:385–400. [DOI] [PubMed] [Google Scholar]
- 5.Lowe GC, Saavedra A, Reed KB, Velazquez AI, Dronca RS, Markovic SN, Lohse CM, Brewer JD. Increasing incidence of melanoma among middle-aged adults: an epidemiologic study in Olmsted County, Minnesota. Mayo Clin Proc 2014;89:52–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Volkovova K, Bilanicova D, Bartonova A, Letasiova S, Dusinska M. Associations between environmental factors and incidence of cutaneous melanoma. Environ Health 2012;11: Suppl 1:S12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jensen JD, Wing GJ, Dellavalle RP. Nutrition and melanoma prevention. Clin Dermatol 2010;28:644–9. [DOI] [PubMed] [Google Scholar]
- 8.Tong LX, Young LC. Nutrition: The future of melanoma prevention? J Am Acad Dermatol 2014;71:151–60. [DOI] [PubMed] [Google Scholar]
- 9.Murzaku EC, Bronsnick T, Rao BK. Diet in dermatology: Part II. Melanoma, chronic urticaria, and psoriasis. J Am Acad Dermatol 2014;71:1053 e1–e16. [DOI] [PubMed]
- 10.Feskanich D, Willett WC, Hunter DJ, Colditz GA. Dietary intakes of vitamins A, C, and E and risk of melanoma in two cohorts of women. Br J Cancer 2003;88:1381–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Naldi L, Gallus S, Tavani A, Imberti GL, La Vecchia C. Risk of melanoma and vitamin A, coffee and alcohol: a case-control study from Italy. Eur J Cancer Prev 2004;13:503–8. [DOI] [PubMed] [Google Scholar]
- 12.Fortes C, Mastroeni S, Melchi F, Pilla MA, Antonelli G, Camaioni D, Alotto M, Pasquini P. A protective effect of the Mediterranean diet for cutaneous melanoma. Int J Epidemiol 2008;37:1018–29. [DOI] [PubMed] [Google Scholar]
- 13.Vinceti M, Malagoli C, Fiorentini C, Longo C, Crespi CM, Albertini G, Ricci C, Lanzoni A, Reggiani M, Virgili A, et al. Inverse association between dietary vitamin D and risk of cutaneous melanoma in a northern Italy population. Nutr Cancer 2011;63:506–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fortes C, Mastroeni S, Boffetta P, Antonelli G, Pilla MA, Botta G, Anzidei P, Venanzetti F. The protective effect of coffee consumption on cutaneous melanoma risk and the role of GSTM1 and GSTT1 polymorphisms. Cancer Causes Control 2013;24:1779–87. [DOI] [PubMed] [Google Scholar]
- 15.Malavolti M, Malagoli C, Fiorentini C, Longo C, Farnetani F, Ricci C, Albertini G, Lanzoni A, Reggiani C, Virgili A, et al. Association between dietary vitamin C and risk of cutaneous melanoma in a population of northern Italy. Int J Vitam Nutr Res 2013;83:291–8. [DOI] [PubMed] [Google Scholar]
- 16.Loftfield E, Freedman ND, Graubard BI, Hollenbeck AR, Shebl FM, Mayne ST, Sinha R. Coffee drinking and cutaneous melanoma risk in the NIH-AARP diet and health study. J Natl Cancer Inst 2015;107: in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Veierød MB, Thelle DS, Laake P. Diet and risk of cutaneous malignant melanoma: a prospective study of 50,757 Norwegian men and women. Int J Cancer 1997;71:600–4. [DOI] [PubMed] [Google Scholar]
- 18.Vinceti M, Rothman KJ, Bergomi M, Borciani N, Serra L, Vivoli G. Excess melanoma incidence in a cohort exposed to high levels of environmental selenium. Cancer Epidemiol Biomarkers Prev 1998;7:853–6. [PubMed] [Google Scholar]
- 19.Vinceti M, Pellacani G, Malagoli C, Bassissi S, Sieri S, Bonvicini F, Krogh V, Seidenari S. A population-based case-control study of diet and melanoma risk in northern Italy. Public Health Nutr 2005;8:1307–14. [DOI] [PubMed] [Google Scholar]
- 20.Vinceti M, Bonvicini F, Pellacani G, Sieri S, Malagoli C, Giusti F, Krogh V, Bergomi M, Seidenari S. Food intake and risk of cutaneous melanoma in an Italian population. Eur J Clin Nutr 2008;62:1351–4. [DOI] [PubMed] [Google Scholar]
- 21.Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol 2002;13:3–9. [DOI] [PubMed] [Google Scholar]
- 22.Bertuccio P, Rosato V, Andreano A, Ferraroni M, Decarli A, Edefonti V, La Vecchia C. Dietary patterns and gastric cancer risk: a systematic review and meta-analysis. Ann Oncol 2013;24:1450–8. [DOI] [PubMed] [Google Scholar]
- 23.Lai JS, Hiles S, Bisquera A, Hure AJ, McEvoy M, Attia J. A systematic review and meta-analysis of dietary patterns and depression in community-dwelling adults. Am J Clin Nutr 2014;99:181–97. [DOI] [PubMed] [Google Scholar]
- 24.De Stefani E, Boffetta P, Correa P, Deneo-Pellegrini H, Ronco AL, Acosta G, Mendilaharsu M. Dietary patterns and risk of cancers of the upper aerodigestive tract: a factor analysis in Uruguay. Nutr Cancer 2013;65:384–9. [DOI] [PubMed] [Google Scholar]
- 25.Rees K, Hartley L, Flowers N, Clarke A, Hooper L, Thorogood M, Stranges S. ‘Mediterranean’ dietary pattern for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev 2013;8:CD009825. [DOI] [PubMed] [Google Scholar]
- 26.Mourouti N, Kontogianni MD, Papavagelis C, Plytzanopoulou P, Vassilakou T, Malamos N, Linos A, Panagiotakos DB. Adherence to the Mediterranean Diet is associated with lower likelihood of breast cancer: A case-control study. Nutr Cancer 2014;66:810–7. [DOI] [PubMed] [Google Scholar]
- 27.Grosso G, Biondi A, Galvano F, Mistretta A, Marventano S, Buscemi S, Drago F, Basile F. Factors associated with colorectal cancer in the context of the Mediterranean diet: a case-control study. Nutr Cancer 2014;66:558–65. [DOI] [PubMed] [Google Scholar]
- 28.Liu X, Wang X, Lin S, Yuan J, Yu IT. Dietary patterns and oesophageal squamous cell carcinoma: a systematic review and meta-analysis. Br J Cancer 2014;110:2785–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Schulze MB, Hoffmann K. Methodological approaches to study dietary patterns in relation to risk of coronary heart disease and stroke. Br J Nutr 2006;95:860–9. [DOI] [PubMed] [Google Scholar]
- 30.Gamba CS, Stefanick ML, Shikany JM, Larson J, Linos E, Sims ST, Marshall J, Van Horn L, Zeitouni N, Tang JY. Low-fat diet and skin cancer risk: the women's health initiative randomized controlled dietary modification trial. Cancer Epidemiol Biomarkers Prev 2013;22:1509–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Guenther PM, Casavale KO, Reedy J, Kirkpatrick SI, Hiza HA, Kuczynski KJ, Kahle LL, Krebs-Smith SM. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet 2013;113:569–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM, Bray GA, Vogt TM, Cutler JA, Windhauser MM, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. N Engl J Med 1997;336:1117–24. [DOI] [PubMed] [Google Scholar]
- 33.Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003;348:2599–608. [DOI] [PubMed] [Google Scholar]
- 34.Agnoli C, Grioni S, Sieri S, Palli D, Masala G, Sacerdote C, Vineis P, Tumino R, Giurdanella MC, Pala V, et al. Italian Mediterranean Index and risk of colorectal cancer in the Italian section of the EPIC cohort. Int J Cancer 2013;132:1404–11. [DOI] [PubMed] [Google Scholar]
- 35.Agnoli C, Krogh V, Grioni S, Sieri S, Palli D, Masala G, Sacerdote C, Vineis P, Tumino R, Frasca G, et al. A priori-defined dietary patterns are associated with reduced risk of stroke in a large Italian cohort. J Nutr 2011;141:1552–8. [DOI] [PubMed] [Google Scholar]
- 36.Reedy J, Krebs-Smith SM, Miller PE, Liese AD, Kahle LL, Park Y, Subar AF. Higher diet quality is associated with decreased risk of all-cause, cardiovascular disease, and cancer mortality among older adults. J Nutr 2014;144:881–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Schwingshackl L, Hoffmann G. Adherence to Mediterranean diet and risk of cancer: a systematic review and meta-analysis of observational studies. Int J Cancer 2014;135:1884–97. [DOI] [PubMed] [Google Scholar]
- 38.Schwingshackl L, Hoffmann G. Diet quality as assessed by the Healthy Eating Index, the Alternate Healthy Eating Index, the Dietary Approaches to Stop Hypertension Score, and health outcomes: A systematic review and meta-analysis of cohort studies. J Acad Nutr Diet 2015;115:780–800. [DOI] [PubMed] [Google Scholar]
- 39.AIOM-AIRTUM. [Cancer figures in Italy 2014]. Brescia (Italy): Intermedia Editore, 2014 (in Italian). [Google Scholar]
- 40.Forsea AM, Del Marmol V, de Vries E, Bailey EE, Geller AC. Melanoma incidence and mortality in Europe: new estimates, persistent disparities. Br J Dermatol 2012;167:1124–30. [DOI] [PubMed] [Google Scholar]
- 41.Pisani P, Faggiano F, Krogh V, Palli D, Vineis P, Berrino F. Relative validity and reproducibility of a food frequency dietary questionnaire for use in the Italian EPIC centres. Int J Epidemiol 1997;26: Suppl 1:S152–60. [DOI] [PubMed] [Google Scholar]
- 42.Pasanisi P, Berrino F, Bellati C, Sieri S, Krogh V. Validity of the Italian EPIC questionnaire to assess past diet. IARC Sci Publ 2002;156:41–4. [PubMed] [Google Scholar]
- 43.Pala V, Sieri C, Palli D, Salvini S, Berrino F, Bellegotti M, Frasca G, Tumino R, Sacerdote C, Fiorini L, et al. Diet in the Italian EPIC cohorts: presentation of data and methodological issues. Tumori 2003;89:594–607. [DOI] [PubMed] [Google Scholar]
- 44.Salvini S, Parpinel M, Gnagnarella P. Banca dati di composizione degli alimenti per studi epidemiologici in Italia. [Food composition database for epidemiological studies in Italy.] 269 ed. Milan: European Institute of Oncology, 1998 (in Italian). [Google Scholar]
- 45.George SM, Ballard-Barbash R, Manson JE, Reedy J, Shikany JM, Subar AF, Tinker LF, Vitolins M, Neuhouser ML. Comparing indices of diet quality with chronic disease mortality risk in postmenopausal women in the Women's Health Initiative Observational Study: evidence to inform national dietary guidance. Am J Epidemiol 2014;180:616–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Guenther PM, Kirkpatrick SI, Reedy J, Krebs-Smith SM, Buckman DW, Dodd KW, Casavale KO, Carroll RJ. The healthy eating index-2010 is a valid and reliable measure of diet quality according to the 2010 dietary guidelines for Americans. J Nutr 2014;144:399–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Obarzanek E, Sacks FM, Vollmer WM, Bray GA, Miller ER 3rd, Lin PH, Karanja NM, Most-Windhauser MM, Moore TJ, Swain JF, et al. Effects on blood lipids of a blood pressure-lowering diet: the Dietary Approaches to Stop Hypertension (DASH) Trial. Am J Clin Nutr 2001;74:80–9. [DOI] [PubMed] [Google Scholar]
- 48.Wengreen H, Munger RG, Cutler A, Quach A, Bowles A, Corcoran C, Tschanz JT, Norton MC, Welsh-Bohmer KA. Prospective study of Dietary Approaches to Stop Hypertension- and Mediterranean-style dietary patterns and age-related cognitive change: the Cache County Study on Memory, Health and Aging. Am J Clin Nutr 2013;98:1263–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Williams CM, Lovegrove JA, Griffin BA. Dietary patterns and cardiovascular disease. Proc Nutr Soc 2013;72:407–11. [DOI] [PubMed] [Google Scholar]
- 50.Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu FB. Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women. Arch Intern Med 2008;168:713–20. [DOI] [PubMed] [Google Scholar]
- 51.Panico S, Mattiello A, Panico C, Chiodini P. Mediterranean dietary pattern and chronic diseases. Cancer Treat Res 2014;159:69–81. [DOI] [PubMed] [Google Scholar]
- 52.Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 1988;124:869–71. [DOI] [PubMed] [Google Scholar]
- 53.Rothman KJ. Six persistent research misconceptions. J Gen Intern Med 2014;29:1060–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Rothman KJ, Greenland S, Lash TL. Modern Epidemiology. Philadelphia: Lippincott Williams & Wilkins, 2012. [Google Scholar]
- 55.Kvaskoff M, Bijon A, Mesrine S, Boutron-Ruault MC, Clavel-Chapelon F. Cutaneous melanoma and endogenous hormonal factors: a large French prospective study. Am J Epidemiol 2011;173:1192–202. [DOI] [PubMed] [Google Scholar]
- 56.Sergentanis TN, Antoniadis AG, Gogas HJ, Antonopoulos CN, Adami HO, Ekbom A, Petridou ET. Obesity and risk of malignant melanoma: a meta-analysis of cohort and case-control studies. Eur J Cancer 2013;49:642–57. [DOI] [PubMed] [Google Scholar]
- 57.Kvaskoff M, Bijon A, Mesrine S, Vilier A, Clavel-Chapelon F, Boutron-Ruault MC. Anthropometric features and cutaneous melanoma risk: a prospective cohort study in French women. Cancer Epidemiol 2014;38:357–63. [DOI] [PubMed] [Google Scholar]
- 58.Benetou V, Trichopoulou A, Orfanos P, Naska A, Lagiou P, Boffetta P, Trichopoulos D. Greek Ec. Conformity to traditional Mediterranean diet and cancer incidence: the Greek EPIC cohort. Br J Cancer 2008;99:191–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Naska A, Trichopoulou A. Back to the future: The Mediterranean diet paradigm. Nutr Metab Cardiovasc Dis 2014;24:216–9. [DOI] [PubMed] [Google Scholar]
- 60.Bertoia ML, Triche EW, Michaud DS, Baylin A, Hogan JW, Neuhouser ML, Tinker LF, Van Horn L, Waring ME, Li W, et al. Mediterranean and Dietary Approaches to Stop Hypertension dietary patterns and risk of sudden cardiac death in postmenopausal women. Am J Clin Nutr 2014;99:344–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Thompson CA, Zhang ZF, Arah OA. Competing risk bias to explain the inverse relationship between smoking and malignant melanoma. Eur J Epidemiol 2013;28:557–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Lee TK, MacArthur AC, Gallagher RP, Elwood MJ. Occupational physical activity and risk of malignant melanoma: the Western Canada Melanoma Study. Melanoma Res 2009;19:260–6. [DOI] [PubMed] [Google Scholar]
- 63.Reddy KK, Gilchrest BA. The role of vitamin D in melanoma prevention: evidence and hyperbole. J Am Acad Dermatol 2014;71:1004–5. [DOI] [PubMed] [Google Scholar]