Purpose: Low serum concentrations of antioxidants may be associated with an increased risk of cancer. Based on the accumulated evidence, we hypothesized that retinoids would elevate serum α-tocopherol. This study was designed to determine whether 9-cis-retinoic acid (9-cis-RA), the most common chemopreventive agent, could alter serum α-tocopherol in former smokers. Because hyperlipidemia is a known side effect of retinoids, we also evaluated the association between serum α-tocopherol and lipids in the same population.

Experimental Design: Subjects who had stopped smoking at least 12 months before the study were randomly assigned to receive oral 9-cis-RA or placebo daily for 3 months. Clinical information and blood samples were obtained monthly; serum α-tocopherol concentrations were measured by high-performance liquid chromatography and lipid levels by enzymatic assays before treatment and every month during the treatment.

Results: Of the 149 subjects in the study, 113 completed 3 months of treatment and provided samples for evaluation of serum α-tocopherol. Serum α-tocopherol levels in the 9-cis-RA group (n = 52) were higher after treatment (r = 0.445, P < 0.01) than before. The incidences of grade ≥2 hypertriglyceridemia and hypercholesterolemia were higher in the 9-cis-RA group than in the placebo group (P = 0.0005 and P = 0.01, respectively), but there were no serious complications related to hyperlipidemia.

Conclusions: Treatment of former smokers with 9-cis-RA significantly increased their serum α-tocopherol levels, and this could be a benefit. In addition, serum α-tocopherol could serve as a biomarker for 9-cis-RA treatment.

Because lung cancer is the leading cause of cancer-related death worldwide (1, 2), chemoprevention has become an increasingly important priority in the effort to reduce its incidence (3). Thus far, the most often used chemopreventive agents for aerodigestive tract cancers have been retinoids (4), which bind and transactivate retinoic acid (RA) receptors (RAR) and retinoid X receptors (RXR), both of which belong to the superfamily of steroid nuclear receptors. Among the retinoids, all-trans-RA is selective for RARs, and 9-cis-RA binds both RARs and RXRs (5). In contrast, 13-cis-RA binds to neither receptor type but is thought to bind to the retinoid receptors after intracellular stereoisomerization to all-trans-RA or 9-cis-RA (6).

Several findings have shown the potential of retinoids as cancer chemopreventive agents (7–12). First, 9-cis-RA has shown antiproliferative activity against a broad range of neoplastic cells, including those from prostate cancer (7), breast cancer (8, 9), leukemia and lymphoma (10), lung cancer (11), and head and neck cancer (12). Second, 9-cis-RA had substantial in vivo anticarcinogenic activity in rat mammary glands (13, 14) and rat colons (15). Finally, in patients with a history of cancer of the head and neck region, 13-cis-RA treatment reduced the incidence of second primary tumors and reversed leukoplakia (i.e., premalignant oral lesions; refs. 16, 17).

Despite these promising findings, enthusiasm for the use of retinoids as chemopreventive agents for lung cancer waned after two large randomized clinical trials ended with disappointing results. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study, which tested the efficacy of the antioxidant dietary supplements α-tocopherol and β-carotene, actually showed that β-carotene significantly increased lung cancer incidence and mortality over the levels observed in participants not taking it (18). The detrimental effects of β-carotene were confirmed by the Beta-Carotene and Retinol Efficacy Trial, which revealed a 28% higher rate of lung cancer and 17% higher overall death rate in participants taking β-carotene than in those not taking the supplement (19). However, further analyses showed that the adverse effects of β-carotene were restricted to active smokers in both trials (20). In addition, there was some evidence of a benefit from retinoid treatment in nonsmokers and former smokers (21), suggesting that the response to retinoids differs in current and former smokers. Supporting these findings, 13-cis-RA treatment of current smokers had no effect on bronchial squamous metaplasia, a histologic abnormality associated with smoking, whereas 13-cis-RA treatment inhibited bronchial squamous metaplasia in former smokers (22). Wang et al. (23) reported that cigarette smoke and high doses of β-carotene induced a decrease in RA concentrations and RAR-β expression and an increase in cell proliferation and the expression of activating protein 1 family members (c-Jun and c-Fos) and cyclin D1 in the lung tissue of ferrets. Liu et al. (24) showed that the oxidative metabolites of β-carotene generated by smoke-induced production of cytochrome P450, which interferes with RA metabolism by down-regulating RAR signaling and suppressing RAR-β expression. We recently showed in a chemopreventive trial that 9-cis-RA could benefit former smokers (25) by increasing RAR-β expression and decreasing metaplasia relative to the findings in the control group. These findings suggest that current and former smokers had different responses to retinoids and that retinoids, especially 9-cis-RA, have potential as chemopreventive agents in former smokers. Based on this notion, we sought to identify other benefits of 9-cis-RA for former smokers.

Antioxidants also have potential as chemopreventive agents (26–28), and low serum levels of antioxidants are associated with an increased risk of cancer (29–31). Because a correlation has been observed between serum levels of retinoids and the antioxidant α-tocopherol (32), we hypothesized that serum levels of α-tocopherol would be elevated in people who had been treated with retinoids. To test our hypothesis, we retrospectively analyzed data from a previously published three-arm, randomized, double-blinded, placebo-controlled trial comparing 9-cis-RA, 13-cis-RA plus α-tocopherol, and placebo daily for 3 months (25). Because hyperlipidemia is a common side effect of retinoid treatment (33, 34), we also evaluated possible associations between serum α-tocopherol and lipid levels in the same subjects.

Subjects. The original trial design, the method of determining compliance with the trial protocol and the monitoring of toxic effects of the drugs have been previously described (25). That study was a three-arm, randomized, double-blinded, placebo-controlled trial comparing 9-cis-RA (100 mg), 13-cis-RA (1 mg/kg) plus α-tocopherol (1,200 IU), and placebo daily for 3 months (25). RAR-β was detected in 69.7% of all baseline biopsy samples, and metaplasia was evident in 6.9% of all baseline samples from 240 subjects enrolled between November 1995 and May 2001. RAR-β expression was restored and metaplasia was reduced after treatment in the 9-cis-RA group. After adjustment for years of smoking, packs per day smoked, and metaplasia, treatment with 9-cis-RA but not with 13-cis-RA plus α-tocopherol led to a statistically significant increase in RAR-β expression compared with placebo. All subjects were former smokers, defined as people with a smoking history of at least 20 pack-years who had stopped smoking at least 1 year before entering the study. To be eligible, subjects had to have adequate renal, hematologic, and hepatic function and must not have taken more than 25,000 IU of vitamin A or other retinoids per day within 3 months of study entry. Subjects with a prior smoking-related cancer were eligible if they had been tumor-free for 6 months before enrollment. Subjects were required to abstain from consuming vitamin supplements during the study. Before being randomly assigned to a treatment group, all eligible subjects provided written informed consent. This study was approved by the institutional review board at the University of Texas M.D. Anderson Cancer Center and by the U.S. Department of Health and Human Services. Subjects were seen monthly and evaluated for compliance with the trial protocol and for drug-related toxic effects. Serum cotinine levels were determined at baseline and at 3 and 6 months after treatment initiation to document compliance with smoking abstinence during the trial. The treatment duration of 3 months was chosen on the basis of toxicity data from prior phase I trials that involved 9-cis-RA treatment (35, 36). In the current analysis, we did not include the group treated with 13-cis-RA plus α-tocopherol to avoid possible confounding effects from oral supplementary α-tocopherol.

Specimen Collection. To analyze changes in α-tocopherol levels and toxicity, blood specimens (10 mL) were drawn from each participant at the beginning of the study and then monthly during treatment. Blood was collected in heparinized tubes and transported immediately to the laboratory, where the specimens were separated and processed. Serum was collected after centrifugation of the blood at 1,500 rpm for 10 minutes at room temperature and was stored at −80°C until it was needed for testing.

High-Performance Liquid Chromatography Analysis for α-Tocopherol. Methods for extracting analytes from serum, quality control variables and the high-performance liquid chromatography (HPLC) methods for analyzing α-tocopherol in serum have been published previously (29). Briefly, a hexane extract of 0.4 mL of serum was injected onto a 3-μm C-18 Spherisorb ODS-2 HPLC column (1050 HPLC system; Hewlett-Packard, Avondale, PA) and eluted with an isocratic solvent consisting of 73% acetonitrile, 12% tetrahydrofuran, 8% methanol, 7% water, 0.025% ammonium acetate, and 0.05% diethylamine (v/v) at 1.2 mL/min. α-Tocopherol was detected at 292 nm. The HPLC system was fully automated and equipped with quaternary pumps, an electronic degasser, insulated column housing, an automatic sampler diode array detector, and software to run the system and perform data management. The coefficient of variation for the pooled quality control samples for all of the analytes was ≤10%.

Biochemical Assays. Serum levels of triglycerides and cholesterol were measured with routine enzymatic methods on a random access clinical chemistry system (Dimension, Dade Behring International, Inc., Newark, NJ).

Statistical Analyses. The Wilcoxon rank sum test and Kruskal-Wallis test were used to test the equal medians of continuous variables between two treatment groups. Either the χ2 test or Fisher's exact test was used to test the association between two categorical variables. Because the distributions of α-tocopherol were skewed, the differences in α-tocopherol levels between treatment groups were tested with the Wilcoxon rank sum test. Changes in serum α-tocopherol levels from baseline levels were tested separately at each subsequent patient visit (i.e., at 1-, 2-, 3-, and 6-month visits) and in each group by using the Wilcoxon rank sum test. All statistical tests were two-sided, with a 5% type I error rate. Statistical analysis was done with standard statistical software, including SAS Release 8.1 (SAS Institute, Cary, NC) and S-Plus 2000 (Mathsoft, Inc., Seattle, WA).

Characteristics of Subjects. The characteristics of the eligible subjects have been described in detail previously (25). Two hundred forty subjects were entered into the original clinical trial, of whom 226 were eligible to be randomly assigned into one of the three treatment groups (9-cis-RA, 13-cis-RA plus α-tocopherol, or placebo); 149 of these participants (the subjects of the current analysis) had been randomly assigned to either the placebo or the 9-cis-RA groups, and 113 of them completed 3 months of treatment. The characteristics of the evaluable subjects are summarized in Table 1. Each treatment group was well balanced for age, sex, race, history of smoking-related cancer, number of pack-years, number of years since stopping smoking, body mass index, and serum levels of cholesterol and triglycerides. Serum cotinine levels drawn at registration, at 3 and 6 months showed that >95% of the participants had serum levels in the range of nonsmokers (<1.0 ng/mL) or passive smokers (1-20 ng/mL) at all three measurement times.

Table 1

Baseline characteristics of subjects according to treatment group

Placebo (n = 61)9-cis-RA (n = 52)P*
Gender (%)    
    Male 37 (60.7) 27 (51.9) 0.35 
    Female 24 (39.3) 25 (48.1)  
Race (%)    
    White 52 (85.3) 49 (94.2) 0.35 
    African 6 (9.8) 2 (3.9)  
    Hispanic 3 (4.9) 1 (1.9)  
    Oriental  
Smoking-related cancer (%)    
    No 54 (88.5) 48 (92.3) 0.54 
    Yes 7 (11.5) 4 (7.7)  
Age    
    Mean ± SD 58.1 ± 8.9 55.7 ± 9.2 0.12 
    Median (range) 58.8 (34.9-73.6) 54.9 (35.9-74.5)  
Body mass index    
    Mean ± SD 27.8 ± 4.1 28.1 ± 5.38 0.92 
    Median (range) 27.1 (20.6-39.4) 27.3 (19.4-44.4)  
Smoking years    
    Mean ± SD 29.1 ± 9.8 27.3 ± 9.5 0.39 
    Median (range) 30 (15-50) 26 (10-50)  
Packs per day    
    Mean ± SD 1.7 ± 0.7 1.9 ± 0.8 0.23 
    Median (range) 1.5 (1-4) 2 (0.8-4)  
Pack-years    
    Mean ± SD 50.2 ± 27.2 52.6 ± 30.5 0.94 
    Median (range) 42.5 (20-135) 42 (20-136)  
Smoking quit years    
    Mean ± SD 10.4 ± 8.8 11.0 ± 8.7 0.56 
    Median (range) 10.1 (1.1-35.2) 7.8 (1.0-38.2)  
Cholesterol    
    Mean ± SD 205.8 ± 37.5 206.9 ± 31.9 0.74 
    Median (range) 198 (141-292) 205 (151-283)  
Triglyceride    
    Mean ± SD 142.4 ± 65.9 152.2 ± 60.4 0.39 
    Median (range) 132 (43-309) 136 (58-282)  
Placebo (n = 61)9-cis-RA (n = 52)P*
Gender (%)    
    Male 37 (60.7) 27 (51.9) 0.35 
    Female 24 (39.3) 25 (48.1)  
Race (%)    
    White 52 (85.3) 49 (94.2) 0.35 
    African 6 (9.8) 2 (3.9)  
    Hispanic 3 (4.9) 1 (1.9)  
    Oriental  
Smoking-related cancer (%)    
    No 54 (88.5) 48 (92.3) 0.54 
    Yes 7 (11.5) 4 (7.7)  
Age    
    Mean ± SD 58.1 ± 8.9 55.7 ± 9.2 0.12 
    Median (range) 58.8 (34.9-73.6) 54.9 (35.9-74.5)  
Body mass index    
    Mean ± SD 27.8 ± 4.1 28.1 ± 5.38 0.92 
    Median (range) 27.1 (20.6-39.4) 27.3 (19.4-44.4)  
Smoking years    
    Mean ± SD 29.1 ± 9.8 27.3 ± 9.5 0.39 
    Median (range) 30 (15-50) 26 (10-50)  
Packs per day    
    Mean ± SD 1.7 ± 0.7 1.9 ± 0.8 0.23 
    Median (range) 1.5 (1-4) 2 (0.8-4)  
Pack-years    
    Mean ± SD 50.2 ± 27.2 52.6 ± 30.5 0.94 
    Median (range) 42.5 (20-135) 42 (20-136)  
Smoking quit years    
    Mean ± SD 10.4 ± 8.8 11.0 ± 8.7 0.56 
    Median (range) 10.1 (1.1-35.2) 7.8 (1.0-38.2)  
Cholesterol    
    Mean ± SD 205.8 ± 37.5 206.9 ± 31.9 0.74 
    Median (range) 198 (141-292) 205 (151-283)  
Triglyceride    
    Mean ± SD 142.4 ± 65.9 152.2 ± 60.4 0.39 
    Median (range) 132 (43-309) 136 (58-282)  
*

The Wilcoxon rank sum test was performed to test of continuous variables between two treatment groups. The χ2 test (for sex) and Fisher's exact test (for race and smoking-related cancer) were performed to test the association between two categorical variables.

Effects of 9-cis-RA Treatment on Serum α-Tocopherol Level. We investigated whether 9-cis-RA treatment affected the serum α-tocopherol level. We found that baseline serum α-tocopherol level of the placebo group was not different from that of 9-cis-RA treatment group (Table 2). Serum α-tocopherol concentrations significantly increased over time in the 9-cis-RA group but not in the placebo group, peaking 1 month after the start of treatment and maintaining this level throughout treatment. This finding indicates that 9-cis-RA can increase serum α-tocopherol concentrations in former smokers.

Table 2

α-Tocopherol level at baseline and during treatment of 9-cis-RA

α-Tocopherol level, median (range, μg/dl)
Placebo (n = 61)9-cis-RA (n = 52)P
Baseline 13,017 (2,241-45,620) 14,002 (6,162-33,779) 0.37* 
1st month 11,818 (1,311-22,306) 16,456 (7,883-48,877) 0.005 
2nd month 11,454 (5,770-26,506) 15,847 (7,607-46,132) 0.003 
3rd month 12,354 (6,788-23,150) 15,381 (8,461-49,383) 0.01 
α-Tocopherol level, median (range, μg/dl)
Placebo (n = 61)9-cis-RA (n = 52)P
Baseline 13,017 (2,241-45,620) 14,002 (6,162-33,779) 0.37* 
1st month 11,818 (1,311-22,306) 16,456 (7,883-48,877) 0.005 
2nd month 11,454 (5,770-26,506) 15,847 (7,607-46,132) 0.003 
3rd month 12,354 (6,788-23,150) 15,381 (8,461-49,383) 0.01 
*

P was obtained from the Wilcoxon rank sum test.

P was obtained from repeated-measures analysis using a mixed model.

To determine whether subject characteristics may have affected their baseline serum α-tocopherol level, we evaluated possible associations between the characteristics listed in Table 1 and baseline serum α-tocopherol levels. Gender, race, body mass index, number of pack-years, number of years since stopping smoking, and levels of cholesterol were not associated with the baseline serum α-tocopherol level (Table 3). However, the baseline serum triglyceride level and age were significantly associated with higher baseline serum α-tocopherol levels. When adjusted for body mass index, the effects of both age (P = 0.02) and baseline serum triglyceride level (P = 0.008) on the baseline serum α-tocopherol concentration were found to be significant by regression analysis.

Table 3

Association between baseline α-tocopherol level and demographic characteristics

Characteristicsα-Tocopherol level, mean ± SD (range, ng/mL)P*
Gender   
    Male 15,090 ± 6,967.4 (7,086-45,620) 0.96 
    Female 14,399 ± 5,993.7 (2,241-33,779)  
Race   
    White 15,097 ± 6,701.4 (2,241-45,620) 0.24 
    African 12,796 ± 5,252.6 (7,086-21,245)  
    Hispanic 11,047 ± 2,203.1 (8,484-13,792)  
Age (y)   
    <60 13,885 ± 6,287.0 (5,575-45,620) 0.02 
    ≥60 16,380 ± 6,758.4 (2,241-33,586)  
Body mass index   
    Nonobese (<28) 14,040 ± 5,036.2 (5,575-29,889) 0.47 
    Obese (≥28) 15,531 ± 6,844.1 (6,162-33,779)  
Pack-years   
    <40 14,357 ± 6,742.6 (5,575-45,620) 0.47 
    ≥40 15,135 ± 6,415.0 (2,241-33,779)  
Smoking quit years   
    <10 13,827 ± 5,515.5 (2,241-30,742) 0.15 
    ≥10 15,921 ± 7,472.0 (5,575-45,620)  
Cholesterol (mg/dl)   
    <200 14,060 ± 5,560.5 (5,575-33,586) 0.32 
    ≥200 15,419 ± 6,992.6 (2,241-45,620)  
Triglyceride (mg/dl)   
    ≤135 12,562 ± 4,550.0 (2,241-25,603) 0.008 
    >135 16,440 ± 7,146.4 (7,681-45,620)  
Characteristicsα-Tocopherol level, mean ± SD (range, ng/mL)P*
Gender   
    Male 15,090 ± 6,967.4 (7,086-45,620) 0.96 
    Female 14,399 ± 5,993.7 (2,241-33,779)  
Race   
    White 15,097 ± 6,701.4 (2,241-45,620) 0.24 
    African 12,796 ± 5,252.6 (7,086-21,245)  
    Hispanic 11,047 ± 2,203.1 (8,484-13,792)  
Age (y)   
    <60 13,885 ± 6,287.0 (5,575-45,620) 0.02 
    ≥60 16,380 ± 6,758.4 (2,241-33,586)  
Body mass index   
    Nonobese (<28) 14,040 ± 5,036.2 (5,575-29,889) 0.47 
    Obese (≥28) 15,531 ± 6,844.1 (6,162-33,779)  
Pack-years   
    <40 14,357 ± 6,742.6 (5,575-45,620) 0.47 
    ≥40 15,135 ± 6,415.0 (2,241-33,779)  
Smoking quit years   
    <10 13,827 ± 5,515.5 (2,241-30,742) 0.15 
    ≥10 15,921 ± 7,472.0 (5,575-45,620)  
Cholesterol (mg/dl)   
    <200 14,060 ± 5,560.5 (5,575-33,586) 0.32 
    ≥200 15,419 ± 6,992.6 (2,241-45,620)  
Triglyceride (mg/dl)   
    ≤135 12,562 ± 4,550.0 (2,241-25,603) 0.008 
    >135 16,440 ± 7,146.4 (7,681-45,620)  
*

P was obtained from the Wilcoxon rank sum test.

Effects of 9-cis-RA Treatment on Serum Concentrations of Triglycerides and Cholesterol. We tested the serum levels of lipids, including triglycerides and cholesterol, in all participants. During treatment, serum levels of lipids were significantly increased in 9-cis-RA group and decreased when treatment with 9-cis-RA was stopped (Fig 1). In testing the serum levels of lipids, including triglycerides and cholesterol, in all participants, we found no significant difference in the incidence of grade 1 hyperlipidemia between the 9-cis-RA group and the placebo group during treatment (Table 4). However, 9-cis-RA group showed significantly more subjects of grade ≥2 hypertriglyceridemia and hypercholesterolemia than the placebo group (P < 0.0001). Of 52 evaluable subjects in the 9-cis-RA group, 22 (42%, P = 0.0001) and 12 (23%, P = 0.0001) developed grade >2 hypertriglyceridemia and hypercholesterolemia, respectively. One of the participants in the 9-cis-RA group experienced grade 4 hypertriglyceridemia and discontinued treatment. However, no serious complications related to hypertriglyceridemia or hypercholesterolemia, such as cardiovascular events, pancreatitis, or death, were experienced in either group, and the hyperlipidemia disappeared after discontinuation of 9-cis-RA treatment. Finally, we evaluated whether the modulation of serum lipid levels was associated with modulation of serum α-tocopherol levels during 3 months of 9-cis-RA treatment. The modulation of serum α-tocopherol levels in the 9-cis-RA treatment group was significantly associated with the changes in serum levels of triglyceride (Fig. 2A) and cholesterol (Fig. 2B).

Fig 1

Effect of 9-cis-RA on the serum levels of (A) triglycerides and (B) cholesterol after 3 months of treatment.

Fig 1

Effect of 9-cis-RA on the serum levels of (A) triglycerides and (B) cholesterol after 3 months of treatment.

Close modal
Table 4

The incidence of hypertriglyceridemia or hypercholesterolemia according to treatment group

National Cancer Institute Common Toxicity CriteriaPlacebo (n = 61)9-cis-RA (n = 52)P*
Hypertriglyceridemia    
    Grade 1 (>2.5 × ULN) 28 23 <0.0001 
    Grade 2 (>2.5-5.0 × ULN) 15  
    Grade 3 (>5.0-10 × ULN)  
    Grade 4 (>10 × ULN)  
Hypercholesterolemia    
    Grade 1 (>ULN-300 mg/dl) 22 24 <0.0001 
    Grade 2 (>300-400 mg/dl)  
    Grade 3 (>400-500 mg/dl)  
    Grade 4 (>500 mg/dl)  
National Cancer Institute Common Toxicity CriteriaPlacebo (n = 61)9-cis-RA (n = 52)P*
Hypertriglyceridemia    
    Grade 1 (>2.5 × ULN) 28 23 <0.0001 
    Grade 2 (>2.5-5.0 × ULN) 15  
    Grade 3 (>5.0-10 × ULN)  
    Grade 4 (>10 × ULN)  
Hypercholesterolemia    
    Grade 1 (>ULN-300 mg/dl) 22 24 <0.0001 
    Grade 2 (>300-400 mg/dl)  
    Grade 3 (>400-500 mg/dl)  
    Grade 4 (>500 mg/dl)  

Abbreviation: ULN, upper limit of normal range.

*

P value was obtained from the Fisher's exact test.

Fig. 2

The correlation between changes in serum levels of lipids and changes in serum levels of α-tocopherol level in the 9-cis-RA and placebo treatment groups. Wilcoxon rank sum test was performed to determine the statistical significance of the changes in the median α-tocopherol level (3-month value minus baseline value) compared with the changes in the median (A) triglyceride (3-month value minus baseline value) (r = 0.68, P < 0.001) and (B) cholesterol (3-month value minus baseline value) (r = 0.5, P = 0.004) levels.

Fig. 2

The correlation between changes in serum levels of lipids and changes in serum levels of α-tocopherol level in the 9-cis-RA and placebo treatment groups. Wilcoxon rank sum test was performed to determine the statistical significance of the changes in the median α-tocopherol level (3-month value minus baseline value) compared with the changes in the median (A) triglyceride (3-month value minus baseline value) (r = 0.68, P < 0.001) and (B) cholesterol (3-month value minus baseline value) (r = 0.5, P = 0.004) levels.

Close modal

To our knowledge, this is the first report showing that 9-cis-RA affects serum α-tocopherol levels and that serum α-tocopherol levels correlated with serum lipid levels in the setting of a chemoprevention trial. Specifically, daily doses of 9-cis-RA increased serum concentrations of α-tocopherol, a well-known antioxidant, in former smokers who had not smoked for at least 1 year.

It has been suggested that an imbalance between oxidants and antioxidants results in a chronic state of oxidative stress that could contribute to various human diseases, including cancer (37). In fact, reactive oxygen species are constantly generated by ionizing and UV radiation, activation of chemical carcinogens, and the presence of heavy metal carcinogens, and these reactive oxygen species can damage DNA and thus cause mutations that can lead to cancers (38). Antioxidants protect cells from DNA damage by directly removing these reactive oxygen species thus reducing DNA damage and potentially decreasing tumorigenesis. Indeed, several studies have already shown the cancer chemopreventive properties of antioxidants (39). For example, dietary supplementation with α-tocopherol has been shown to prevent exercise-induced DNA damage (40), and results from a randomized phase II chemoprevention trial showed that α-tocopherol decreased oxidative DNA damage (41). The use of α-tocopherol to prevent human cancer has been evaluated for lung cancer, oral leukoplakia, colorectal polyps, and prostate cancer (30, 39, 41, 42). In a separate study, vitamin C, vitamin E, and β-carotene supplementation also significantly reduced endogenous oxidative DNA damage in lymphocytes and increased resistance to oxidative damage induced by hydrogen peroxide (43). Taken together, these findings suggest that 9-cis-RA can induce potential cancer chemopreventive activities in former smokers by increasing α-tocopherol levels.

We found age to be significantly associated with higher baseline serum α-tocopherol levels, an observation also found in other cohort studies (30, 44, 45). The positive association between α-tocopherol levels and age can be explained by age-related changes in the metabolism and transport of α-tocopherol; for example, the activity of lipoprotein lipase, an enzyme that releases α-tocopherol from chylomicrons and transfers it to tissues, has been shown to decrease with age (30). We also noted a positive relationship between baseline triglyceride and α-tocopherol levels. Because α-tocopherol is preferentially bound by the hepatic tocopherol binding protein, which is incorporated into both low- and high-density lipoproteins (44, 46), increases in serum triglyceride levels could be expected in association with increased serum α-tocopherol levels. Baseline serum cholesterol levels also seemed associated with baseline serum α-tocopherol levels, but this apparent relationship was not statistically significant, perhaps because baseline serum cholesterol levels were distributed over a narrow range. Because hyperlipidemia, especially hypertriglyceridemia, is a well-known effect of the clinical use of retinoids, we evaluated serum lipid concentrations and found that hypertriglyceridemia and hypercholesterolemia developed in 42% and 23% of patients, all former smokers, treated with 9-cis-RA, respectively, proportions that are consistent with previous findings (47, 48). Moreover, elevation of serum α-tocopherol concentrations was significantly associated with elevated levels of lipids in this group. A plausible explanation for this finding is the unique binding characteristics of 9-cis-RA to retinoid receptors that have transcription activities. Several studies have provided evidence that retinoids induce hyperlipidemia by activating RAR, RXR, or both (49–51). In one study, the simultaneous activation of RAR and RXR by panagonists induced 2- to 3-fold higher levels of serum triglycerides than did activation of RAR alone (51). Vu-Dac et al. (52) reported that RXR activation increased the expression of apo C-III, a key player in plasma triglyceride metabolism. Furthermore, 9-cis-RA induces the expression of the ATP-binding cassette transporter A1, a major regulator of peripheral cholesterol efflux and plasma high-density lipoprotein metabolism and increases apo A-I-mediated cholesterol efflux, thereby increasing the cholesterol concentration in the blood (53, 54). These data indicate that RAR and RXR ligands can act synergistically to induce high serum concentrations of lipids, suggesting that hyperlipidemia may be a greater issue for 9-cis-RA, an RAR/RXR panagonist, than for other RAs (6).

The risk of hyperlipidemia resulting from retinoid treatment in our chemoprevention trial was not clear. More small dense low-density lipoprotein particles are present in the blood in hyperlipidemia, and these particles are more susceptible to oxidation and are associated with enhanced peroxidation products (55). A high serum level of α-tocopherol, which breaks the chain reaction of lipid peroxidation by donating a hydrogen atom to the reactive oxygen species, could result in the formation of a relatively stable α-tocopherol radical that is thought to be recycled by ascorbate and ubiquinol (56, 57), suggesting that the ability of 9-cis-RA to modulate serum α-tocopherol concentrations may protect cells from increased concentrations of lipid peroxidants. Indeed, no serious complications related to hypertriglyceridemia or hypercholesterolemia, such as cardiovascular problems, pancreatitis, or death, were experienced in the subjects treated with 9-cis-RA in our study. Only one patient with grade 4 hypertriglyceridemia stopped 9-cis-RA treatment, and even that patient had no hyperlipidemia-associated complications. Furthermore, measurements of serum lipid levels after treatment was discontinued showed normalization of lipid levels in all subjects. However, considering recent findings that the risk of lung cancer is higher in recent former smokers than in current smokers (58), the inclusion of subjects (albeit few) with prior cancers and recent smoking cessation might have clouded our data. Therefore, cognizance of the potential cardiovascular risk should be well advised when oral supplementation of agents that increase serum α-tocopherol levels is considered for chemoprevention trial in a group of former smokers. Close monitoring of symptoms and signs of hyperlipidemia and the use of lipid- and cholesterol-lowering agents would be recommended for subjects taking 9-cis-RA for long periods. In addition, further study is needed to clarify the relationship between hyperlipidemia and retinoid treatment and to define the adequate duration of retinoid treatment in a chemoprevention trial.

In summary, to our knowledge, our results show for the first time that 9-cis-RA treatment increased serum α-tocopherol levels in former smokers. From the perspective of directing future public health initiatives, our current findings are important because increasing serum levels of the antioxidant may inhibit cell proliferation and angiogenesis and stimulate apoptosis and immune function (59–61). Additional work is warranted to determine whether other antioxidants, such as vitamin C, vitamin A, and β-carotene, are regulated by 9-cis-RA treatment in former smokers and whether the changes in α-tocopherol levels induced by 9-cis-RA correlate with their ability to reduce the risk of lung cancer in former smokers.

Grant Support: NIH grants U19 CA68437 (W. Hong) and 1R01 CA100816-01A1 (H.-Y. Lee), American Cancer Society grant RSG-04-082-01-TBE (H.-Y. Lee), and Department of Defense grants DAMD17-01-1-0689 and W81XWH-04-1-0142-01-VITAL (W. Hong).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Stewart AK, Bland KI, McGinnis LS Jr, et al. Clinical highlights from the National Cancer Data Base, 2000.
CA Cancer J Clin
2000
;
50
:
171
–83.
2
Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of 25 major cancers in 1990.
Int J Cancer
1999
;
80
:
827
–41.
3
Tsao AS, Kim ES, Hong WK. Chemoprevention of cancer.
CA Cancer J Clin
2004
;
54
:
150
–80.
4
Lippman SM, Lee JJ, Karp DD, et al. Randomized phase III intergroup trial of isotretinoin to prevent second primary tumors in stage I non-small-cell lung cancer.
J Natl Cancer Inst
2001
;
93
:
605
–18.
5
Sun SY, Lotan R. Retinoids and their receptors in cancer development and chemoprevention.
Crit Rev Oncol Hematol
2002
;
41
:
41
–55.
6
Marchetti MN, Sampol E, Bun H, et al. In vitro metabolism of three major isomers of retinoic acid in rats. Intersex and interstrain comparison.
Drug Metab Dispos
1997
;
25
:
637
–46.
7
Blutt SE, Allegretto EA, Pike JW, et al. 1,25-Dihydroxyvitamin D3 and 9-cis-retinoic acid act synergistically to inhibit the growth of LNCaP prostate cells and cause accumulation of cells in G1.
Endocrinology
1997
;
138
:
1491
–7.
8
Rubin M, Fenig E, Rosenauer A, et al. 9-cis-Retinoic acid inhibits growth of breast cancer cells and down-regulates estrogen receptor RNA and protein.
Cancer Res
1994
;
54
:
6549
–56.
9
Gottardis MM, Lamph WW, Shalinsky DR, et al. The efficacy of 9-cis-retinoic acid in experimental models of cancer.
Breast Cancer Res Treat
1996
;
38
:
85
–96.
10
Lutzky J, Vujicic M, Yamanishi DT, et al. Antiproliferative effects of all-trans-retinoic acid (tRA) and 9-cis-retinoic acid (9-cisRA) on human lymphoid cell lines abstract.
Proceedings of the American Association for Cancer Res
1993
;
34
:
292
.
11
Guzey M, Demirpence E, Criss W, et al. Effects of retinoic acid (all-trans and 9-cis) on tumor progression in small-cell lung carcinoma.
Biochem Biophys Res Commun
1998
;
242
:
369
–75.
12
Giannini F, Maestro R, Vukosavljevic T, et al. All-trans, 13-cis, and 9-cis retinoic acids induce a fully reversible growth inhibition in HNSCC cell lines: implications for in vivo retinoic acid use.
Int J Cancer
1997
;
70
:
194
–200.
13
Anzano MA, Byers SW, Smith JM, et al. Prevention of breast cancer in the rat with 9-cis-retinoic acid as a single agent and in combination with tamoxifen.
Cancer Res
1994
;
54
:
4614
–7.
14
Anzano MA, Peer CW, Smith JM, et al. Chemoprevention of mammary carcinogenesis in the rat: combined use of raloxifene and 9-cis-retinoic acid.
J Natl Cancer Inst
1996
;
88
:
123
–5.
15
Zheng Y, Kramer PM, Olson G, et al. Prevention by retinoids of azoxymethane-induced tumors and aberrant crypt foci and their modulation of cell proliferation in the colon of rats.
Carcinogenesis
1997
;
18
:
2119
–25.
16
Hong WK, Sporn MB. Recent advances in chemoprevention of cancer.
Science
1997
;
278
:
1073
–7.
17
Hong WK, Endicott J, Itri LM, et al. 13-cis-retinoic acid in the treatment of oral leukoplakia.
N Eng J Med
1986
;
315
:
1501
–5.
18
The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and β carotene on the incidence of lung cancer and other cancers in male smokers.
N Engl J Med
1994
;
330
:
1029
–35.
19
Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of β carotene and vitamin A on lung cancer and cardiovascular disease.
N Engl J Med
1996
;
334
:
1150
–5.
20
Albanes D, Heinonen OP, Taylor PR, et al. α-tocopherol and β-carotene supplements and lung cancer incidence in the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study: effects of base-line characteristics and study compliance.
J Natl Cancer Inst
1996
;
88
:
1560
–70.
21
Lippman SM, Lee JJ, Karp DD, et al. Randomized phase III intergroup trial of isotretinoin to prevent second primary tumors in stage I non-small-cell lung cancer.
J Natl Cancer Inst
2001
;
93
:
605
–18.
22
Lee JS, Lippman SM, Benner SE, et al. Randomized placebo-controlled trial of isotretinoin in chemoprevention of bronchial squamous metaplasia.
J Clin Oncol
1994
;
12
:
937
–45.
23
Wang XD, Liu C, Bronson RT, et al. Retinoid signaling and activator protein-1 expression in ferrets given β-carotene supplements and exposed to tobacco smoke.
J Natl Cancer Inst
1999
;
91
:
60
–6.
24
Liu C, Russell RM, Wang XD. Exposing ferrets to cigarette smoke and a pharmacological dose of β-carotene supplementation enhance in vitro retinoic acid catabolism in lungs via induction of cytochrome P450 enzymes.
J Nutr
2003
;
133
:
173
–9.
25
Kurie JM, Lotan R, Lee JJ, et al. Treatment of former smokers with 9-cis-retinoic acid reverses loss of retinoic acid receptor-β expression in the bronchial epithelium: results from a randomized placebo-controlled trial.
J Natl Cancer Inst
2003
;
95
:
206
–14.
26
Sawant SS, Kandarkar SV. Role of vitamins C and E as chemopreventive agents in the hamster cheek pouch treated with the oral carcinogen-DMBA.
Oral Dis
2000
;
6
:
241
–7.
27
Trickler D, Shklar G. Prevention by vitamin E of experimental oral carcinogenesis.
J Natl Cancer Inst
1987
;
78
:
165
–9.
28
Ross D, Moldeus P. Antioxidant defense system and oxidative stress. In: Vigo-Pelfrey C, editors. Membrane lipid oxidation. Boca Raton (FL): CRC Press; 1990. p. 151–70.
29
Goodman GE, Schaffer S, Omenn GS, Chen C, King I. The association between lung and prostate cancer risk, and serum micronutrients: results and lessons learned from β-carotene and retinol efficacy trial.
Cancer Epidemiol Biomarkers Prev
2003
;
12
:
518
–26.
30
Woodson K, Tangrea JA, Barrett MJ, Virtamo J, Taylor PR, Albanes D. Serum α-tocopherol and subsequent risk of lung cancer among male smokers.
J Natl Cancer Inst
1999
;
91
:
1738
–43.
31
Taylor PR, Qiao YL, Abnet CC. Prospective study of serum vitamin E levels and esophageal and gastric cancers.
J Natl Cancer Inst
2003 Sep 17
;
95
:
1414
–6.
32
Basile G, Gangemi S, Lo Balbo C, et al. Correlation between serum retinol and α-tocopherol levels in centenarians.
J Nutr Sci Vitaminol (Tokyo)
2003
;
49
:
287
–8.
33
Bershad S, Rubinstein A, Paterniti JR, et al. Changes in plasma lipids and lipoproteins during isotretinoin therapy for acne.
N Engl J Med
1985
;
313
:
981
–5.
34
Lowe NJ, David M. New retinoids for dermatologic diseases. Uses and toxicity.
Dermatol Clin
1988
;
6
:
539
–52.
35
Kurie JM, Lee JS, Griffin T, et al. Phase I trial of 9-cis retinoic acid in adults with solid tumors.
Clin Cancer Res
1996
;
2
:
287
–93.
36
Miller VA, Rigas JR, Benedetti FM, et al. Initial clinical trial of the retinoid receptor pan agonist 9-cis retinoic acid.
Clin Cancer Res
1996
;
2
:
471
–5.
37
Hu JJ, Chi CX, Frenkel K, et al. α-tocopherol dietary supplement decreases titers of antibody against 5-hydroxymethyl-2′-deoxyuridine (HMdU).
Cancer Epidemiol Biomarkers Prev
1999
;
8
:
693
–8.
38
Frenkel K. Role of active oxygen species in carcinogenesis. In: Bertino JR, editor. Encyclopedia of cancer. San Diego (CA): Academic Press; 1997. p. 233–45.
39
Benner SE, Wargovich MJ, Lippman SM, et al. Reduction in oral mucosa micronuclei frequency following α-tocopherol treatment of oral leukoplakia.
Cancer Epidemiol Biomarkers Prev
1994
;
3
:
73
–6.
40
Hartmann A, Niess AM, Grunert-Fuchs M, et al. Vitamin E prevents exercise-induced DNA damage.
Mutat Res
1995
;
346
:
195
–202.
41
McKeown-Eyssen G, Holloway C, Jazmaji V, et al. A randomized trial of vitamins C and E in the prevention of recurrence of colorectal polyps.
Cancer Res
1988
;
48
:
4701
–5.
42
Heinonen O, Albanes D, Virtamo J, et al. Prostate cancer and supplementation with α-tocopherol and β-carotene: incidence and mortality in a controlled trial.
J Natl Cancer Inst
1998
;
90
:
440
–6.
43
Duthie SJ, Ma A, Ross MA, Collins A. Antioxidant supplementation decreases oxidative DNA damage in human lymphocytes.
Cancer Res
1996
;
56
:
1291
–5.
44
White E, Kristal AR, Shikany JM, et al. Correlates of serum α- and γ-tocopherol in the Women's Health Initiative.
Ann Epidemiol
2001
;
11
:
136
–44.
45
Papas A, Stacewicz-Sapuntzakis M, Lagiou P, Bamia C, Chloptsios Y, Trichopoulou A. Plasma retinol and tocopherol levels in relation to demographic, lifestyle and nutritional factors of plant origin in Greece.
Br J Nutr
2003
;
89
:
83
–7.
46
Traber MG, Sies H. Vitamin E in humans: demand and delivery.
Annu Rev Nutr
1996
;
16
:
321
–47.
47
Bershad S, Rubinstein A, Paterniti JR, et al. Changes in plasma lipids and lipoproteins during isotretinoin therapy for acne.
N Engl J Med
1985
;
313
:
981
–5.
48
Lowe NJ, David M. New retinoids for dermatologic diseases. Uses and toxicity.
Dermatol Clin
1988
;
6
:
539
–52.
49
Standeven AM, Beard RL, Johnson AT, et al. Retinoid-induced hypertriglyceridemia in rats is mediated by retinoic acid receptors.
Fundam Appl Toxicol
1996
;
33
:
264
–71.
50
Rigas JR, Miller VA, Levine DM, et al. Lipoprotein alterations in patients treated with novel retinoids [abstract]. Proceedings of the American Association of Cancer Research 1995;36.
51
Standeven AM, Thacher SM, Yuan YD, et al. Retinoid X receptor agonist elevation of serum triglycerides in rats by potentiation of retinoic acid receptor agonist induction or by action as single agents.
Biochem Pharmacol
2001
;
62
:
1501
–9.
52
Vu-Dac N, Gervois P, Torra IP, et al. Retinoids increase human apo C-III expression at the transcriptional level via the retinoid X receptor: contribution to the hypertriglyceridemic action of retinoids.
J Clin Invest
1998
;
102
:
625
–32.
53
Koldamova RP, Lefterov IM, Ikonomovic MD, et al. 22R-Hydroxycholesterol and 9-cis-retinoic acid induce ATP-binding cassette transporter A1 expression and cholesterol efflux in brain cells and decrease amyloid β secretion.
J Biol Chem
2003
;
278
:
13244
–56.
54
Mascrez B, Ghyselinck NB, Watanabe M, et al. Ligand-dependent contribution of RXRβ to cholesterol homeostasis in Sertoli cells.
EMBO Rep
2004
;
5
:
285
–90.
55
de Graaf J, Hak-Lemmers HL, Hectors MP, et al. Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein subfraction in healthy subjects.
Arterioscler Thromb
1991
;
11
:
298
–306.
56
Oostenbrug GS, Mensink RP, Al MD, et al. Maternal and neonatal plasma antioxidant levels in normal pregnancy, and the relationship with fatty acid unsaturation.
Br J Nutr
1998
;
80
:
67
–73.
57
De Vriese SR, Dhont M, Christophe AB. Oxidative stability of low density lipoproteins and vitamin E levels increase in maternal blood during normal pregnancy.
Lipids
2001
;
36
:
361
–6.
58
Ebbert JO, Yang P, Vachon CM, et al. Lung cancer risk reduction after smoking cessation: observations from a prospective cohort of women. J Clin Oncol 2003;921–6.
59
Sigounas G, Anagnostou A, Steiner M. dl-α-Tocopherol induces apoptosis in erythroleukemia, prostate, and breast cancer cells.
Nutr Cancer
1997
;
28
:
30
–5.
60
Shklar G, Schwartz JL. Vitamin E inhibits experimental carcinogenesis and tumour angiogenesis.
Eur J Cancer B Oral Oncol
1996
;
32B
:
114
–9.
61
Meydani SN, Beharka AA. Recent developments in vitamin E and immune response.
Nutr Rev
1996
;
56
:
S49
–58.