Table 1.

Characteristics of studies included in the review.

Reference	Country	Subject Characteristics	Study Design	Outcome	Supplement Intake	Study Duration/Follow-Up	Main Reported Results
Sluyter et al. [35]	New Zealand	Total subjects = 442 (vitamin D group = 226; placebo = 216) Never-smokers (vitamin D group = 122; placebo = 103) Ever-smokers (current/former) (vitamin D group = 104; placebo = 113) Ever-smokers+ vitamin D deficient (vitamin D group = 26; placebo = 28) Ever-smokers+ asthma/COPD (vitamin D group = 25; placebo = 35) Age = 50–84 years	RDBPC	Lung function, asthma/COPD	Vitamin D3 capsules 200,000 IU at baseline, followed by monthly 100,000 IU or monthly placebo soft-gel oral capsules	1 year	Across ever-smoker subgroups, the effect of vitamin D supplementation compared to placebo was significant. Ever-smokers (β = 57 mL, P = 0.03), ever-smokers with vitamin D deficiency (β = 122 mL, P = 0.04), ever-smokers with asthma/COPD (β = 160 mL, P = 0.004). A significant improvement in lung function was observed among ever-smokers with asthma/COPD over non-smokers (P = 0.0005).
Satia et al. [25]	US	Total subjects = 77,126 Lung cancer cases = 521 (75% were non-small lung cancer), non-smokers = 42, former smokers (quit ≥10 years) = 226, former smokers (quit <10 years) = 93, current smokers = 155 Non-cancer controls = 76,605 Age = 50–76 years	Prospective cohort study	LC risk	Lutein, retinol, vitamin A, β-carotene, and lycopene	10 years	Use of retinol supplement was associated with a statistically significant increased risk for former smokers (quit ≥ 10 years) (HR = 1.46, 95% CI = 0.93 to 2.29). None of the other supplements were associated with cancer risk in current/former smokers
Slatore et al. [26]	US	Total subjects = 77,126 Lung cancer cases = 521, non-smokers = 42, former smokers (quit ≥10 years) = 226, former smokers (quit <10 years) = 93, current smokers = 155 Non-cancer controls = 76,605 Age = 50–76 years	Prospective cohort study	LC risk	Multivitamins, folate, vitamin C, and E	10 years	Use of vitamin E supplement was associated with a statistically significant increased risk for current smokers (HR = 1.59, 99% CI = 1.50 to 2.41). All other supplements were not associated with a reduced or increased risk in current/former smokers.
Brasky et al. [27]	US	Total subjects = 77,118 Lung cancer cases = 808, non-smokers = 60, former smokers (quit >10 years) = 334, former smokers (quit <10 years) = 152, current smokers = 251 Non-cancer controls = 76,310 Age = 50–76 years	Prospective cohort study	LC risk	Vitamin B6, B12, and B9/folic acid	10 years	Use of supplemental B6 (HR = 2.93, 95% CI = 1.50 to 5.72) and B12 (HR = 3.71, 95% CI = 1.77 to 7.74) was associated with LC risk among current smokers only. In contrast, use of vitamin B9 was not associated with LC risk among current/former smokers.
Wu et al. [32]	China	Total subjects = 72,829 female non-smokers Age = 40–70 years	Prospective cohort study	LC risk	Vitamin E	12 years	Vitamin E supplement was associated with LC risk among females likely exposed to side-stream smoke (HR = 2.06, 95% CI = 1.31 to 3.23).
Takata et al. [33]	China	Total subjects = 71,267 female non-smokers Age = 40–70 years	Prospective cohort study	LC risk	Calcium/vitamin D	12 years	No association was observed between calcium/vitamin D supplement and LC risk among female who had passive smoking exposure at home/work.
Skeie et al. [31]	Norway	Total subjects = 2997 female cancer patients with solid tumors Lung cancer patients = 217 Non-smokers = 3.9% Former smokers = 12.6% Current smokers = 83.5% Age = mean ~58 years	Prospective cohort study	LC mortality	Multivitamins, ginseng/Q10, herbs/plants, minerals, vitamin B, C, and E	3 years	Whole year daily use of cod liver oil (RR = 0.56, 95% CI = 0.35 to 0.92) and daily (RR = 0.70, 95% CI = 0.49 to 0.99) and occasional (RR = 0.55, 95% CI = 0.31 to 0.97) users of other dietary supplements were associated with a statistically significant decreased death.
Mahabir et al. [41]	US	Total subjects = 482,875 men and women Non-smokers = 170,401 Former smokers = 237,216 Current smokers = 57,142 Age = 50–71 years	Prospective cohort study	LC risk	Calcium, magnesium, zinc, iron, selenium, copper	7 years	Mineral supplements were not associated with risk in current/former smokers.
Park et al. [42]	US	Total subjects = 182,099 men and women Former smokers (supplement users—men) = 51.9% Former smokers (supplement users—women) = 29.5% current smokers (supplement users—men) = 16.5% Current smokers (supplement users—women) = 13.2% Age = 45–75 years	Prospective cohort study	LC risk and mortality	Multivitamins/mineral	11 years	Multivitamin/mineral supplements were not associated with mortality and LC risk in current/former smokers.
Tao et al. [43]	US	Total subjects = 36,382 postmenopausal women(CaD group = 18,176; placebo = 18,106) Non-smokers (CaD group = 9325; placebo = 9428) Former smokers (CaD group = 7255; placebo = 7133) Current smokers (CaD group = 1405; placebo = 1356) Age = 50–79 years	RDBPC	LC risk	Daily dose of vitamin D (400 IU D3) plus calcium (1000 mg calcium carbonate) or daily placebo multivitamin tablets	11 years	After the follow-up period, the CaD supplementation was not associated with LC risk among current/former smokers. However, the CaD supplementation increases the risk among current smokers with a high Ca:Mg ratio (> 2.53) (HR = 1.36, 95% CI = 0.78 to 2.36).
Duffield-Lillico et al. [44]	US	Total subjects = 1312 cancer patients Lung cancer patients = 60 (selenium group = 25; placebo = 35) Non-smokers (selenium group = 34; placebo = 30) Former smokers (selenium group = 39; placebo = 40) Current smokers (selenium group = 27; placebo = 30) Age = mean ~61 years	RDBPC	LC risk	Daily dose of 200 µg selenium or a placebo	13 years	Selenium supplementation reduced LC risk among current/former, although the reduction was not statistically significant (RR = 0.74, 95% CI = 0.44 to 1.24).
Tanvetyanon and Bepler [45]	US and Finland	Total subjects = 109,394 lung cancer patients Former smokers = 15,076 Current smokers = 24,109	Systematic review/meta-analysis of four RDBPC	LC risk	Daily dose of 20–30 mg β-carotene or a placebo	2–12 years across studies	β-carotene supplementation increased LC risk among current smokers but not former smokers (OR = 1.24, 95% CI = 1.10 to 1.39).
Albanes et al. [28]	Finland	Total subjects = 29,133 male current smokers Age = 50–69 years	RDBPC	LC risk	α-tocopherol acetate (50 mg/day), β-carotene (20 mg/day), both β-carotene and α-tocopherol, or placebo	5–8 years	β-carotene supplementation increased LC risk (RR = 1.25, 95% CI = 1.07 to 1.46). However, α-tocopherol had no effect.
Virtamo et al. [29]	Finland	Total subjects = 25,563 male current smokers (≥18 cigarettes/day) Age = 50–69 years	RDBPC	LC risk/mortality	α-tocopherol acetate (50 mg/day), β-carotene (20 mg/day), both β-carotene and α-tocopherol, or placebo	18 years	β-carotene and α-tocopherol supplementation had no effect on LC risk/mortality.
Middha et al. [46]	Finland	Total subjects = 29,133 male current smokers β-carotene supplementation group: Ultra-light cigarettes = 1359, light = 2224, regular = 9565,nonfiltered = 1421 Ventilated filtered = 3639, unventilated filtered = 9509, no filter = 1412 Age = 50–69 years	RDBPC	LC risk	β-carotene (20 mg/day) or placebo	5–8 years	β-carotene supplementation increased LC risk among male current smokers, regardless of tar/nicotine content of cigarettes smoked. Ultra-light = (HR = 1.31, 95% CI = 1.91 to 1.89) Light = (HR = 1.18, 95% CI = 0.89 to 1.57) Regular = (HR = 1.15, 95% CI = 1.01 to 1.31) Non-filtered = (HR = 1.22, 95% CI = 0.91 to 1.64) Ventilated filtered = (HR = 1.23, 95% CI = 0.98 to 1.54) Unventilated filtered = (HR = 1.15, 95% CI = 1.01 to 1.31) No filter = (HR = 1.22, 95% CI = 0.91 to 1.64)
Van Zandwijk et al. [47]	Europe (not specified)	Total subjects = 2592 cancer patients Lung cancer patients = 1023 (40% with non-small cell lung cancer) Non-smokers = 168 Current/former = smokers = 2405 Age = mean ~61 years	RCT	LC risk	N-acetylcysteine (600 mg/day for two years), retinyl palmitate (300,000 IU/day for the 1st year plus 150,000 IU/day for the 2nd year), both N-acetylcysteine and retinyl palmitate or no treatment	2 years	N-acetylcysteine and retinyl palmitate supplements were not found to be effective in reducing LC risk among smokers.
Cheng et al. [48]	US	Total subjects = 128,779 postmenopausal women Lung cancer patients = 1771 Age = 50–79 years	RDBPC	LC risk	Daily 400 IU of vitamin D3 and 1 g of Ca or placebo	7 years	CaD supplementation reduced LC risk among women, although the reduction was not statistically significant (HR = 0.87, 95% CI = 0.70 to 1.07). The CaD supplementation increases the risk among current smokers with a high total vitamin A intake (≥3000 µg/day RAE) (HR = 2.26, 95% CI = 1.02 to 5.01).
Cheng et al. [50]	US	Total subjects = 596 postmenopausal women 298 lung cancer cases vs. 298 control non-smokers Age = 50–70 years	Case-control study	LC risk	Calcium/vitamin D	1 year	Non-smoker women with high serum 25(OH)D concentrations at baseline and exposure to the CaD trial intervention were associated with a low risk of LC (OR = 0.42, 95% CI = 0.16 to 1.14).
Cheng et al. [49]	US	Total subjects = 1428 men and women 749 lung cancer cases and 679 non-cancer controls Former smokers = 222 Current smokers = 527 Age = 50–69 years	Case-cohort design	LC risk	CARET active intervention: 25,000 IU retinyl palmitate plus 30 mg β-carotene/day	17 years	Former smokers with high vitamin D intake and received the CARET active intervention were associated with a low risk of LC (HR = 0.25, 95% CI = 0.08 to 0.76).
Goodman et al. [30]	US	Total subjects = 13,447 lung cancer patients CARET intervention group = 6902, placebo group = 6545 Former smokers = 6447 Current smokers = 7000 Age = 50–69 years	RCT	LC risk/mortality	25,000 IU retinyl palmitate plus 30 mg β-carotene/day or placebo	6 years	Current smokers had a lower LC risk (1.22 vs. 1.42; RR = 1.22, 95% CI = 0.98 to 1.51) and a lower mortality rate (1.27 vs. 1.66; RR = 1.27, 95% CI = 0.99 to 1.64) than those in the trial phase. Former smokers had an increased risk (1.11 vs. 0.80; RR = 1.11, 95% CI = 0.85 to 1.47) and a lower mortality rate (1.12 vs. 1.27; RR = 1.12, 95% CI = 0.83 to 1.52) than those in the trial phase.

Abbreviation: RDBPC, randomized double blind placebo control; RCT, randomized control trial; LC, lung cancer; HR, hazard ratio; RR, relative risk; OR, odds ratio; CI, confidence interval.