Scientific opinion on the tolerable upper intake level for selenium

View full-text article in PMC

. 2023 Jan 20;21(1):e07704. doi: 10.2903/j.efsa.2023.7704

Cohort name Country Reference Follow‐up Funding	Original Cohort (N total) Exclusion criteria Study population (n, sex and age at baseline^(a))	Ascertainment of outcome	Exposure groups^(a) n	Incident cases	Model covariates	Results
HPFS & NHS USA Matthews et al. (2019) HPFS: 26 yr NHS: 28 yr Prospective cohort Public	HPFS N = 51,529 NHS N = 121,700 Population sampled: male health professionals (HPFS) and registered nurses (NHS) Excluded: prior history of cancer, non‐Caucasian, no toenail clippings HPFS n = 3,730 NHS n = 6,708 HPFS Sex: males NHS Sex: females Ethnicity: Caucasian HPFS Age (yr): 40–75 NHS Age (yr): 30–55	Incident BCC, SCC and MSC identified through self‐administered biennial questionnaires; confirmed on the basis of medical and pathology reports for SCC and melanoma. Self‐reported BCC not confirmed against medical report (validity previously demonstrated, with 96% and 84% of cases, in females and males, respectively, confirmed by pathology records)	Toenail Se (μg/g, median) HPFS/NHS Q1(ref): 0.67/0.63 Q2: 0.77/0.71 Q3: 0.84/0.77 Q4: 0.94/0.83 Q5: 1.17/0.95 n per quintile NR	HPFS/NHS BCC Q1(ref): 158/292 Q2: 178/329 Q3: 180/310 Q4: 180/313 Q5: 184/309 SCC Q1(ref): 32/43 Q2: 34/47 Q3: 48/53 Q4: 37/40 MSC Q1(ref): 16/12 Q2: 18/17 Q3: 20/9 Q4: 19/19	Assay batch, age (in years), number of severe sunburns, number of moles, hair colour, family history of melanoma, history of SCC or melanoma, UV exposure at residence (quintiles), history of physical examination, and sun reactions (none, burn, painful burn/blisters) (and Fitzpatrick score (I‐VI) for females)	HR (95% CI) for BCC Q1(ref): 1 Q2: 1.06 (0.93, 1.20) Q3: 1.00 (0.88, 1.13) Q4: 0.99 (0.87, 1.13) Q5: 0.99 (0.87, 1.13) p _trend 0.64 HR (95% CI) for SCC* Q1(ref): 1 Q2: 1.06 (0.77, 1.45) Q3: 1.27 (0.93, 1.72) Q4: 0.90 (0.65, 1.26) p _trend 0.69 HR (95% CI) for MSC* Q1(ref): 1 Q2: 1.22 (0.74, 2.03) Q3: 0.99 (0.58, 1.68) Q4: 1.34 (0.80, 2.24) p _trend 0.43 *Se intake divided in quartiles due to smaller sample sizes
The Nambour Skin Cancer Study Australia van der Pols et al. (2009) 8 yr Prospective cohort Public	N = 1,621 Population sampled: Adult residents of Nambour, sub‐tropical area of Australia Excluded: skin cancer at baseline; no serum Se measurement n = 485 Sex (% F): 54 Ethnicity: Caucasian Age (yr): 20–69	BCC and SCC identified through biannual questionnaires; cases confirmed through histologic reports.	*Serum Se (μmol/L; median (min‐max))** T1: 0.9 (0.4–1.0) T2: 1.1 (1.1–1.2) T3: 1.4 (1.3–2.8) 1 μmol/L = 79 μg/L BCC, n per tertile* T1: 163 T2: 159 T3:163 SCC, n per tertile T1: 162 T2: 158 T3: 165	BCC: T1: 20 T2: 22 T3: 17 SCC: T1: 17 T2: 21 T3: 21	Model 1: age and sex Model 2: Model 1 + pack‐years of smoking, alcohol intake (continuous), time spent outdoors on weekdays, and history of skin cancer before baseline	RR (95% CI) for BCC Model 1 T1(ref): 1 T2: 1.09 (0.56, 2.10) T3: 0.57 (0.28, 1.17) p _trend 0.14 Model 2 T1(ref): 1 T2: 1.02 (0.56, 1.87) T3: 0.43 (0.21, 0.86) p _trend 0.02 RR (95% CI) for SCC Model 1 T1(ref): 1 T2: 0.97 (0.48, 1.95) T3: 0.44 (0.19, 1.00) p _trend 0.06 Model 2 T1(ref): 1 T2: 0.86 (0.44, 1.67) T3: 0.36 (0.15, 0.82) p _trend 0.02
The Nambour Skin Cancer Study Australia Heinen et al. (2007) 8 yr Prospective cohort Public	N = 1,621 Population sampled: Adult residents of Nambour, sub‐tropical area of Australia Excluded: skin cancer at baseline; incomplete FFQ, energy intakes outside the normal ranges n = 1,001 Sex (% F): 54 Ethnicity: Caucasian Age (yr): 20–69	BCC and SCC identified through biannual questionnaires; cases confirmed through histologic reports.	Se intake assessed through a SFFQ Se intake (μg/d; median (min‐max)) T1: 70.1 (34.3–76.2) T2: 82.2 (76.2–89.3) T3: 99.1 (89.3–168.9) BCC/SCC, n per tertile NR	BCC: T1: 84 T2: 122 T3: 115 SCC: T1: 61 T2: 63 T3: 97	Model 1: age, sex, treatment allocation during the Nambour trial of b‐carotene and/or sunscreen (prior to baseline) Model 2: age, sex, energy intake (kJ/day), skin colour, tanning ability of skin, elastosis of the neck, number of painful sunburns, smoking, treatment allocation, use of dietary supplements (yes/no), history of skin cancer before baseline	RR (95% CI) for BCC Model 1 T1(ref): 1 T2: 1.2 (0.71, 1.9) T3: 1.0 (0.62, 1.7) p _trend 0.90 Model 2 T1(ref): 1 T2: 1.2 (0.73, 1.9) T3: 0.95 (0.59, 1.5) p _trend 0.81 RR (95% CI) for SCC Model 1 T1(ref): 1 T2: 0.75 (0.43, 1.3) T3: 1.1 (0.63, 1.8) p _trend 0.73 Model 2 T1(ref): 1 T2: 1.1 (0.59, 1.9) T3: 1.3 (0.77, 2.3) p _trend 0.28
VITAL cohort USA Asgari et al. (2009) 5 yr (mean) Prospective cohort	N = 77,719 Population sampled: Adult residents of western Washington Excluded: MSC at baseline, non‐Caucasian or did not report their race n = 69,671 Sex (% F): 54 Ethnicity: Caucasian Age (yr): 50–76	Incident cases of MSC identified through linkage with Surveillance, Epidemiology, and End Results (SEER) cancer registry; included melanoma in situ, malignant melanoma NOS, superficial spreading melanoma, lentigo maligna melanoma, nodular melanoma, and other subtypes including melanoma within a junctional nevus, spindle cell melanoma, acral lentiginous melanoma, and desmoplastic melanoma.	Self‐reported Se intake from food supplements (μg/d) over the 10 y prior to baseline n, per category: G1, none: 23,855 G2, > 0–≤ 20: 28,613 G3, > 20–< 50: 11,414 G4: ≥ 50: 5,392	G1: 152 G2: 195 G3: 76 G4: 37	Age, gender, education (high school or less, some college, advanced degree), 1st degree family history melanoma (no, yes), personal history of NMSC (no, yes), ever had moles removed (no, yes), freckles between ages 10–20 yr (no, yes), had ≥ 3 severe sunburns between ages 10–20 years (no, yes), natural red/blond hair between ages 10–20 yr (no, yes), and reaction to 1‐h in strong sunlight	RR (95% CI) for MSC G1(ref): 1 G2: 1.09 (0.88, 1.36) G3: 1.01 (0.77, 1.33) G4: 0.98 (0.69, 1.41) p _trend 0.98
Skin Cancer Prevention Study USA Karagas et al. (1997) 5 yr (mean) Nested case–control Public	N = 1,805 Population sampled: Adults participating to the Skin Cancer Prevention study (at least one BCC or SCC cancer previously removed) Excluded: NR n = 392 Cases: 132 Controls: 264 (matched for age, sex and study center) Sex (% F): 11 Ethnicity: Caucasian Age (yr): 35–84	SCC cases detected at annual examinations by study dermatologists; biopsy specimens read locally and reviewed centrally by the study dermatopathologist. Skin lesions removed at times other than the scheduled study examinations identified through patient questionnaires and confirmed against medical records.	Plasma Se (ppm, mean (SE)) Cases: 0.127 (0.0018) Controls: 0.128 (0.0012) n, per quartile for first SCC/any SCC: Q1 ≤ 0.12: 111/118 Q2 0.121–0.130: 73/84 Q3 0.131–0.140: 81/91 Q4 > 0.140: 84/99	first SCC/any SCC Q1: 49/49 Q2: 18/19 Q3: 24/28 Q4: 28/35	Smoking	OR (95% CI) for first SCC Q1(ref): 1 Q2: 0.45 (0.23, 0.87) Q3: 0.62 (0.31, 1.24) Q4: 0.67 (0.35, 1.29) p _trend 0.25 OR (95% CI) for any SCC Q1(ref): 1 Q2: 0.44 (0.23, 0.84) Q3: 0.74 (0.38, 1.44) Q4: 0.86 (0.47, 1.58) p _trend 0.89
MCHES Finland Knekt et al. (1990); Knekt et al. (1991) 10 yr (median) Nested case–control Public	N = 39,268 Population sampled: general population, aged 15–99 yr Excluded: history of cancer n = 378 Cases: 126 Controls: 252 (matched for sex, age and municipality) Sex (% F): 50 Ethnicity: Caucasian Age (yr): 15–99	BCC and MSC cases identified from the nationwide Finnish Cancer Registry	Serum Se (μg/L) BCC cases: Males: 60.7 ± 16.9 Females: 62.7 ± 15.3 BCC controls: Males: 61.9 ± 14.5 Females: 62.6 ± 14.8 Quintiles Q1: < 0.49 Q2: 49–57 Q3: 58–66 Q4: 67–77 Q5: ≥ 78 Melanoma cases: 59.8 (NR) Melanoma controls: 61.3 (NR) n per quintile of intake NR	Total BCC cases: 64 males 62 females Melanoma cases: 10 (mixed sex; % NR) Melanoma controls: 18 (mixed sex; % NR) Cases per quintile of intake NR MSC cases = 10	Smoking	RR (95% CI NR) for BCC Males Q1(ref): 1 Q2: 0.94 Q3: 0.38 Q4: 0.41 Q5: 0.54 p _trend 0.429 Females Q1(ref): 1 Q2: 1.81 Q3: 1.63 Q4: 0.94 Q5: 1.55 p _trend 0.743 RR for MSC, per 1 SD increase of serum Se (crude) 0.79 (95% CI NR) p _trend 0.68

Cohort name

Country

Reference

Follow‐up

Funding

Original Cohort (N total)

Exclusion criteria

Study population (n, sex and age at baseline^(a))

Ascertainment of outcome

Exposure groups^(a)

Incident cases

Model covariates

Results

HPFS & NHS

USA

Matthews et al. (2019)

HPFS: 26 yr

NHS: 28 yr

Prospective cohort

Public

HPFS N = 51,529

NHS N = 121,700

Population sampled: male health professionals (HPFS) and registered nurses (NHS)

Excluded: prior history of cancer, non‐Caucasian, no toenail clippings

HPFS n = 3,730

NHS n = 6,708

HPFS Sex: males

NHS Sex: females

Ethnicity: Caucasian

HPFS Age (yr):

40–75

NHS Age (yr):

30–55

Incident BCC, SCC and MSC identified through self‐administered biennial questionnaires; confirmed on the basis of

medical and pathology reports for SCC and melanoma. Self‐reported BCC not confirmed against medical report (validity previously demonstrated, with 96% and 84% of cases, in females and males, respectively, confirmed by pathology records)

Toenail Se (μg/g, median)

HPFS/NHS

Q1(ref): 0.67/0.63

Q2: 0.77/0.71

Q3: 0.84/0.77

Q4: 0.94/0.83

Q5: 1.17/0.95

n per quintile NR

HPFS/NHS

BCC

Q1(ref): 158/292

Q2: 178/329

Q3: 180/310

Q4: 180/313

Q5: 184/309

SCC

Q1(ref): 32/43

Q2: 34/47

Q3: 48/53

Q4: 37/40

MSC

Q1(ref): 16/12

Q2: 18/17

Q3: 20/9

Q4: 19/19

Assay batch, age (in years), number of severe sunburns, number of moles, hair colour, family history of melanoma, history of SCC or melanoma, UV exposure at residence (quintiles), history of

physical examination, and sun reactions (none, burn, painful burn/blisters) (and Fitzpatrick score (I‐VI) for females)

HR (95% CI) for BCC

Q1(ref): 1

Q2: 1.06 (0.93, 1.20)

Q3: 1.00 (0.88, 1.13)

Q4: 0.99 (0.87, 1.13)

Q5: 0.99 (0.87, 1.13)

p _trend 0.64

HR (95% CI) for SCC*

Q1(ref): 1

Q2: 1.06 (0.77, 1.45)

Q3: 1.27 (0.93, 1.72)

Q4: 0.90 (0.65, 1.26)

p _trend 0.69

HR (95% CI) for MSC*

Q1(ref): 1

Q2: 1.22 (0.74, 2.03)

Q3: 0.99 (0.58, 1.68)

Q4: 1.34 (0.80, 2.24)

p _trend 0.43

*Se intake divided in quartiles due to smaller sample sizes

The Nambour Skin Cancer Study

Australia

van der Pols et al. (2009)

8 yr

Prospective cohort

Public

N = 1,621

Population sampled: Adult residents of Nambour, sub‐tropical area of Australia

Excluded: skin cancer at baseline; no serum Se measurement

n = 485

Sex (% F): 54

Ethnicity: Caucasian

Age (yr): 20–69

BCC and SCC identified through biannual questionnaires; cases confirmed through histologic

reports.

Serum Se (μmol/L*; median (min‐max))

T1: 0.9 (0.4–1.0)

T2: 1.1 (1.1–1.2)

T3: 1.4 (1.3–2.8)

*1 μmol/L = 79 μg/L

BCC, n per tertile

T1: 163

T2: 159

T3:163

SCC, n per tertile

T1: 162

T2: 158

T3: 165

BCC:

T1: 20

T2: 22

T3: 17

SCC:

T1: 17

T2: 21

T3: 21

Model 1: age and sex

Model 2: Model 1 + pack‐years of smoking, alcohol intake (continuous), time spent outdoors on weekdays, and history of skin cancer before baseline

RR (95% CI) for BCC

Model 1

T1(ref): 1

T2: 1.09 (0.56, 2.10)

T3: 0.57 (0.28, 1.17)

p _trend 0.14

Model 2

T1(ref): 1

T2: 1.02 (0.56, 1.87)

T3: 0.43 (0.21, 0.86)

p _trend 0.02

RR (95% CI) for SCC

Model 1

T1(ref): 1

T2: 0.97 (0.48, 1.95)

T3: 0.44 (0.19, 1.00)

p _trend 0.06

Model 2

T1(ref): 1

T2: 0.86 (0.44, 1.67)

T3: 0.36 (0.15, 0.82)

p _trend 0.02

The Nambour Skin Cancer Study

Australia

Heinen et al. (2007)

8 yr

Prospective cohort

Public

N = 1,621

Population sampled: Adult residents of Nambour, sub‐tropical area of Australia

Excluded: skin cancer at baseline; incomplete FFQ, energy intakes outside the normal ranges

n = 1,001

Sex (% F): 54

Ethnicity: Caucasian

Age (yr): 20–69

BCC and SCC identified through biannual questionnaires; cases confirmed through histologic

reports.

Se intake assessed through a SFFQ

Se intake (μg/d; median (min‐max))

T1: 70.1 (34.3–76.2)

T2: 82.2 (76.2–89.3)

T3: 99.1 (89.3–168.9)

BCC/SCC, n per tertile NR

BCC:

T1: 84

T2: 122

T3: 115

SCC:

T1: 61

T2: 63

T3: 97

Model 1: age, sex, treatment allocation during the Nambour trial of b‐carotene and/or sunscreen (prior to baseline)

Model 2: age, sex, energy intake (kJ/day), skin colour, tanning ability of skin, elastosis of the neck, number of painful sunburns, smoking, treatment allocation, use of dietary supplements (yes/no), history of skin cancer before baseline

RR (95% CI) for BCC

Model 1

T1(ref): 1

T2: 1.2 (0.71, 1.9)

T3: 1.0 (0.62, 1.7)

p _trend 0.90

Model 2

T1(ref): 1

T2: 1.2 (0.73, 1.9)

T3: 0.95 (0.59, 1.5)

p _trend 0.81

RR (95% CI) for SCC

Model 1

T1(ref): 1

T2: 0.75 (0.43, 1.3)

T3: 1.1 (0.63, 1.8)

p _trend 0.73

Model 2

T1(ref): 1

T2: 1.1 (0.59, 1.9)

T3: 1.3 (0.77, 2.3)

p _trend 0.28

VITAL cohort

USA

Asgari et al. (2009)

5 yr (mean)

Prospective cohort

N = 77,719

Population sampled: Adult residents of western Washington

Excluded: MSC at baseline, non‐Caucasian or did not report their race

n = 69,671

Sex (% F): 54

Ethnicity: Caucasian

Age (yr): 50–76

Incident cases of MSC identified through linkage with Surveillance,

Epidemiology, and End Results (SEER) cancer registry; included melanoma in situ, malignant melanoma NOS, superficial spreading melanoma, lentigo maligna melanoma, nodular melanoma, and other subtypes including melanoma within a junctional nevus, spindle cell melanoma, acral

lentiginous melanoma, and desmoplastic melanoma.

Self‐reported Se intake from food supplements (μg/d) over the 10 y prior to baseline

n, per category:

G1, none: 23,855

G2, > 0–≤ 20: 28,613

G3, > 20–< 50: 11,414

G4: ≥ 50: 5,392

G1: 152

G2: 195

G3: 76

G4: 37

Age, gender, education (high school or less, some college, advanced degree), 1st degree family history melanoma (no, yes), personal history of NMSC (no, yes), ever had moles removed (no, yes), freckles between ages 10–20 yr (no, yes), had ≥ 3 severe sunburns between ages 10–20 years (no, yes), natural red/blond hair between ages 10–20 yr (no, yes), and reaction to 1‐h in strong sunlight

RR (95% CI) for MSC

G1(ref): 1

G2: 1.09 (0.88, 1.36)

G3: 1.01 (0.77, 1.33)

G4: 0.98 (0.69, 1.41)

p _trend 0.98

Skin Cancer Prevention Study

USA

Karagas et al. (1997)

5 yr (mean)

Nested case–control

Public

N = 1,805

Population sampled: Adults participating to the Skin Cancer Prevention study (at least one BCC or SCC cancer previously removed)

Excluded: NR

n = 392

Cases: 132

Controls: 264 (matched for age, sex and study center)

Sex (% F): 11

Ethnicity: Caucasian

Age (yr): 35–84

SCC cases detected at annual examinations by study dermatologists; biopsy specimens read locally and reviewed centrally by the study dermatopathologist. Skin lesions removed at times other than the scheduled study examinations identified through patient questionnaires and confirmed against medical records.

Plasma Se (ppm, mean (SE))

Cases: 0.127 (0.0018)

Controls: 0.128 (0.0012)

n, per quartile for first SCC/any SCC:

Q1 ≤ 0.12: 111/118

Q2 0.121–0.130: 73/84

Q3 0.131–0.140: 81/91

Q4 > 0.140: 84/99

first SCC/any SCC

Q1: 49/49

Q2: 18/19

Q3: 24/28

Q4: 28/35

Smoking

OR (95% CI) for first SCC

Q1(ref): 1

Q2: 0.45 (0.23, 0.87)

Q3: 0.62 (0.31, 1.24)

Q4: 0.67 (0.35, 1.29)

p _trend 0.25

OR (95% CI) for any SCC

Q1(ref): 1

Q2: 0.44 (0.23, 0.84)

Q3: 0.74 (0.38, 1.44)

Q4: 0.86 (0.47, 1.58)

p _trend 0.89

MCHES

Finland

Knekt et al. (1990); Knekt et al. (1991)

10 yr (median)

Nested case–control

Public

N = 39,268

Population sampled: general population, aged 15–99 yr

Excluded: history of cancer

n = 378

Cases: 126

Controls: 252 (matched for sex, age and municipality)

Sex (% F): 50

Ethnicity: Caucasian

Age (yr): 15–99

BCC and MSC cases identified from the nationwide Finnish Cancer Registry

Serum Se (μg/L)

BCC cases:

Males: 60.7 ± 16.9

Females: 62.7 ± 15.3

BCC controls:

Males: 61.9 ± 14.5

Females: 62.6 ± 14.8

Quintiles

Q1: < 0.49

Q2: 49–57

Q3: 58–66

Q4: 67–77

Q5: ≥ 78

Melanoma cases: 59.8 (NR)

Melanoma controls: 61.3 (NR)

n per quintile of intake NR

Total BCC cases:

64 males

62 females

Melanoma cases: 10 (mixed sex; % NR)

Melanoma controls: 18 (mixed sex; % NR)

Cases per quintile of intake NR

MSC cases = 10

Smoking

RR (95% CI NR) for BCC

Males

Q1(ref): 1

Q2: 0.94

Q3: 0.38

Q4: 0.41

Q5: 0.54

p _trend 0.429

Females

Q1(ref): 1

Q2: 1.81

Q3: 1.63

Q4: 0.94

Q5: 1.55

p _trend 0.743

RR for MSC, per 1 SD increase of serum Se (crude)

0.79 (95% CI NR)

p _trend 0.68

BCC: basal cell carcinoma; BMI: body mass index; CI: confidence interval; F: females; HR: hazard ratio; HPFS: Health Professionals Follow‐up study; MCHES: Mobile Clinic Health Examination Survey; MSC: melanoma skin cancer; NHS: Nurses' Health Study; NMSC: non‐melanoma skin cancer; NR: not reported; NZ: New Zealand; OR: odds ratio; RR: relative risk; Qx: quintile/quartile x; SCC: squamous cell carcinoma; RR: relative risk; SD: standard deviation; SFFQ: semi‐quantitative food frequency questionnaire; VITAL: VITamins And Lifestyle; yr: year.

(a) Mean ± SD (range), unless specified otherwise.