In Honor of the Conclusion of Breaking Bad; An Overview of the Clinical Research on Nutrition and Lung Cancer Prevention

Perhaps the most integral element of TV show Breaking Bad is Walter White’s lung cancer. The show, which concluding its amazing 5 season run this week, serves as a good opportunity to talk about lung cancer, as well as what the available clinical research suggests as far as preventing it.

Among all types of cancers, lung cancer represents the type generally found to be responsible for the most deaths, regardless of gender or race. Although this is likely due largely to the effect of smoking cigarettes, lung cancer is still one of the most common types of cancer among non-smokers.

Given the disease’s pervasiveness, it would seem immensely valuable to examine how diet and nutrient intake may be able to modulate and reduce the rates of lung cancer incidence and mortality. This review will focus almost entirely on evidence from randomized, controlled trials on humans with lung cancer incidence and/or death as endpoints. Despite their importance, this review will ignore the effect of drugs such as smoked tobacco and aspirin (which by the way show a harmful and protective effect, respectively).

For a TL;DR version skip to the conclusion section.

B-Carotene and Retinol (Vitamin A):

B-carotene, an antioxidant and precursor to retinol (also known as vitamin A), has been frequently associated with a decreased risk of lung cancer in smokers (1) and non smokers (2). When this theory was tested to see if giving people at risk of lung cancer B-carotene supplements reduces their risk of the disease, the opposite effect occurred–their risk increased.

The ATBC trial was a study on male smokers conducted in Finland in which more than 14,000 men took daily 20 mg supplements of B-carotene while an equally sized group of men took a placebo (3). By the end of the trial the group taking B-carotene experienced an 18 percent increased risk of lung cancer and a borderline significant increased risk of death from lung cancer.

The CARET trial was a similar study on 4060 males exposed to asbestos and 14,254 men and women smokers, both risk factors for lung cancer (4). Each group was randomized into two additional groups, with one subgroup getting a supplement with 30 mg of B-carotene and 25,000 IU’s of retinol and the other getting a placebo. By the end of the trial both the asbestos exposed workers and the smokers taking the supplement saw their risk from lung cancer increase. This study leads to two important points:

1. Although we cannot say for sure whether the B-carotene, retinol, or both combined were to blame for the increased lung cancer rates, retinol seems relatively innocuous given that studies show retinol supplements are generally protective or not a risk factors for lung cancer (more on this shortly).

2. Some of the smokers quit smoking before the CARET study started. Those who did experienced a non-significant decreased risk of lung cancer from taking the supplement. This fits with the finding of the Physician’s Health Study, an RCT in which roughly 22,000 male physicians were given a 50 mg B-carotene supplement or a placebo (5). Although not very statistically significant, a 22% decrease in lung cancer rates was observed in non-smokers of the B-carotene arm compared to non-smokers taking a placebo.

Alright, I said we’d look at retinol and lung cancer so let’s dive into that. While the CARET study isn’t consistent with the idea that retinol isn’t dangerous to lung cancer risk, many other studies are. First, a study of roughly 1,000 asbestos exposed workers found that a retinol supplement seemed to decrease lung cancer rates and significantly decreased rates of mesothelioma, a form of lung cancer, compared to B-carotene (6). It should be pointed out that a similar study of retinol versus a placebo in asbestos exposed workers found that a retinol supplement had no effect on rates of mesothelioma (7).

A large trial in China found that a combined supplement of retinol and zinc led to a near significant 18% lower risk of lung cancer compared to a placebo (8), seen in the graph below (the Y-axis being risk of lung cancer, the dashed line represents the supplement group, the dark line the placebo group):

Picture 12

Of course, this beneficial effect could have been due to zinc, which some preliminary observational evidence suggest may help prevent lung cancer (12,13).
Finally, a study on 307 people undergoing treatment for stage I non small-cell lung cancer found that a high dose retinol supplement for a year, compared to no treatment, reduced the rates of new tumors (9). However, a similar study on more than 1000 lung cancer patients found that retinol had no effect (10).

Thus, we are left with a few tentative conclusions on the effect of vitamin A, its precursors, and lung cancer risk.

First, there is strong evidence that people at high risk of lung cancer (heavy smokers and people exposed to asbestos) may see their lung cancer rates increase from taking in large amounts of B-carotene. I say large amounts because the lowest amount of B-carotene shown to increase lung cancer in at risk individuals, 20 mg in the ATBC study, is still equivalent to the amount of B-carotene in 5 and 1/2 pounds of carrots, which would be an impressive quantity to consume unless you have a juicer. And this says nothing of carotene bioavailability, which would likely be lower in carrots than a supplement (11). It is unclear from clinical evidence what effect lower intakes of B-carotene might have.

Second, people who both don’t smoke and aren’t exposed to asbestos have not been shown to have an increased lung cancer risk from such large amounts of B-carotene and weak evidence even suggests B-carotene may decrease their risk.

Third, there is very little evidence to suggest retinol supplementation increases lung cancer risk and some to suggest it does not. Weak evidence suggests it might be helpful.


The NPC4 trial was a study conducted on roughly 1,300 men with a history of skin cancer (14). In the study, taking a daily 200 ug supplement of selenium, compared to a placebo, led to a borderline significant reduction in the rates of lung cancer. Reanalysis revealed that this benefit was almost entirely from a decreased risk in subjects who started the study with low blood levels of selenium (15), suggesting that lung cancer is increased by selenium insufficiency.

The SELECT trial recruited over 17,000 men to take a daily 200 ug selenium supplement and found that it very clearly did not prevent lung cancer compared to a placebo (16). This may have been due to the men in the SELECT trial had higher blood levels of selenium (135 v. 113 ng/ml). In fact, the NPC4 trial even reported that participants with the highest blood selenium levels (yet still with levels below the average levels of subjects in the SELECT trial), saw their risk of cancer non-significantly increase when supplementing selenium.

Thus, it would seem that selenium is only protective when you aren’t getting enough of it in the first place and additionally, too much selenium may itself increase risk.

Finally, the SU.VI.MAX study included 13,000 men and women, giving half a supplement containing vitamin C, vitamin E, beta carotene, selenium, and zinc and the other half a placebo(29). By the end of this study the males (but not females) had a non-significant decrease in the rates of respiratory tract cancers, including lung cancers. While difficult to determine which component(s) of this supplement was responsible for this outcome, the relative failure of vitamins C and B-carotene to reduce lung cancer rates, especially in smokers, suggests selenium might have been beneficial (although the zinc could also have helped). This study is weak evidence on its own, but contributes to the plurality of evidence.

Saturated and Linoleic acid/Omega-6 Fats:

The MRFIT, or Multiple Risk Factors Intervention Trial, was a randomized trial to prevent heart disease in men at risk of the disease (17). Of the 12,866 men recruited for the study, half were either put in an intervention group, the other half a usual care group. The intervention group was counseled to quit smoking and replace food sources of saturated fat with polyunsaturated fat (primarily omega-6 fat rich vegetable oils). The authors reported the following finding after a follow up of the study was performed:

“Participants were randomized either to special intervention (SI), which included an intensive smoking cessation program, or to usual care (UC). After 16 years of follow-up, lung cancer mortality rates were higher in the SI than in the UC group. Since rates of smoking cessation in SI were higher than those for UC for the 6 years of the trial, and since risk of lung cancer mortality is known to decline with smoking cessation, these results were unexpected…From randomization through December 1990, 135 SI and 117 UC participants died from lung cancer”.

Given this, it seems difficult to say which dietary change played what role in this result. Did lung cancer increase because consuming lots of omega-6 increases lung cancer rates? Or is saturated fat and/or cholesterol, both of which the intervention group consumed less of (18), uniquely beneficial in preventing lung cancer? Or might additional compounds the groups consumed different amounts of, like vitamin E in vegetable oils or conjugated linoleic acid in dairy fat, be explanatory?

Evidence from animal experiments suggest omega-6 fats from vegetable oils may promote lung cancer (19-23,26). However, additional animal experiments also support the idea that CLA, found in dairy fat (primarily from grass fed cows), might decrease rates of lung cancer (24,25).

Two weak human studies suggest omega-6 is more important. The Lyon Diet Heart Trial reported non-significantly more lung cancer in a group following a diet lower in both saturated fat and omega-6 fat (27). Given that the other elements of this groups diet that were different (vitamin C, vitamin E, ALA) do not appear to increase lung cancer risk, this would point to the negative effect of O6 theory as being more likely. The Women’s Health Initiative also reported non-significantly less lung cancer in subjects following a diet lower in all major types of fat (28), further suggesting saturated fat may not be a uniquely protective element in lung cancer development,.
Ultimately, it would seem stronger evidence points to omega-6 fatty acid rich vegetable oils increasing lung cancer risk than saturated fats or CLA being protective.

Another study similar to the MRFIT study was called the LA Veterans Administration trial (30). It split 846 men into two groups, with one group given food with saturated fat sources replaced with high omega-6 fat containing vegetable oils (this was a double blind study). By the end of the trial it was observed that rates of lung and bronchus cancer were higher in the group making this dietary change to more omega-6 fat. Interestingly, while the number of smokers in each group was similar, the smokers in the vegetable oil group smoked less cigarettes yet still had more lung cancer.

Just to reiterate: in both of the above trials, groups eating vegetable oil in place of saturated fat still experienced greater lung cancer rates, despite favorable smoking rates in said groups.

So, to recap, decent evidence evidence suggests replacing saturated fat sources with high omega-6 vegetable oils will increase one’s risk of lung cancer. Weak evidence suggests this is due to the inclusion of the O6 vegetable oil itself as the most significant factor responsible for this.

Vitamin D

The hypothesis that Vitamin D is possibly protective against lung cancer has been proposed based on the findings of some observational studies (32-33,35,36) and animal experiments (34,37).

This is supported by the results of one trial on 1,179 older women (38). When 1/3 of the women were given vitamin D in addition to a calcium supplement their rates of lung cancer were non-significantly lower than a comparable group given only calcium.

A large subset study from the Women’s Health Initiative, utilizing over 36,000 women, examined the effect of a combined calcium and vitamin D supplement on risk of various diseases, including cancer (42). It found a non-significant decrease in lung cancer among those taking the supplement. Although consistent with a protective role of vitamin D, the degree to which this can be attributed to the vitamin D rather than the calcium is unclear.

Unfortunately, other studies offering data about vitamin D and lung cancer have not yet been published (e.g. the VIDAL study) or have been too small to interpret conclusively (39).

Thus, weak evidence suggests vitamin D might protect against lung cancer.

Vitamin E

Vitamin E has been tested in a number of studies, allowing us to look at its effect on lung cancer rates. Of note is the fact that every study utilized vitamin E in the form alpha-tocopherol, which may be important.

The SELECT and ATBC trials, each discussed in the sections on selenium and B-carotene respectively, also studied the health effects of vitamin E compared to a placebo. Both trials reported the vitamin did not significant change lung cancer incidence. The same finding was seen in the Physicians Health Study trial on vitamin E (43), in which vitamin E lowered lung cancer rates by a not very significant 11%.

The Women’s Health Study gave almost 20,000 women a vitamin E supplement while another near 20,000 received a placebo (50). The results of this was a not very significant 9% increase in lung cancer in the vitamin E group. Essentially, no effect. Certainly not a positive one.

Finally, the HOPE trial of older, generally health afflicted people had 4,761 take a vitamin E supplement and 4,780 take a placebo (44). Among these subjects, lung cancer risk was decreased by a significant 28%.

So why the differences in study findings? Why did the Women’s Health Study make it seem like vitamin E was such a waste of time for preventing lung cancer? Let’s look to this graph:

Vitamin E child

In the graph above, the Y-axis line at 1 equals no change in risk of lung cancer, with deviations below indicating decreased risk and those above indicating increased risk (I used the risk of dying from lung cancer from the SELECT trial as its endpoint).
As can be observed, the risk of lung cancer seems to decrease until about 400 IU ‘s a day, at which point risk increases, with 600 IU’s bringing about an increased risk. This certainly makes it seem like there is a point where vitamin E becomes less beneficial.

Another study supporting this, albeit very weakly (leading me not to include it in the above graph), is the MRC/BHF trial (57). It found that a vitamin supplement including 600 IU’s of vitamin E as well as vitamin C and B-carotene non-significantly increased lung cancer incidence. This result likely could have been due to chance or B-carotene.

Several theories why vitamin E show potential harm after a certain point have been proposed. Some have suggested vitamin E may deplete retinol levels (45,46), which we have seen may be a protective nutrient against lung cancer. Others have speculated that the alpha-tocopherol form used in these studies may be problematic because it interferes with the gamma tocopherol form (40) (the most common form in food), which may be superior to the alpha form (47-49).

Weak evidence thus suggests a moderate amount of vitamin E might prevent lung cancer, with this effect disappearing and potentially reversing itself at higher doses. Time will hopefully provide conclusive evidence regarding whether the gamma form is superior to the alpha form in this regard.

B Vitamins

Several of the B-vitamins have been tested, sometimes in combination, for their effect on cancer rates in multiple trials.

In the WAFACS trial, a study on about 5,500 women, taking a supplement with folic acid (vitamin B9), vitamin B12, and vitamin B6 had no effect on lung cancer compared to a placebo (51). The similar SEARCH trial on about 12,000 subjects also found no effect of vitamin B12 and folic acid on lung cancer rates (55).

Two studies performed in Norway, the NORVIT and WENBIT trials, tested various vitamin B supplements on disease rates. In a population of about 7,000 people, taking a folic acid supplement was found to increase lung rates by a near significant 59% (52). Lung cancer deaths were also increased by a less significant 53%. Taking vitamin B6, on the other hand, produced no apparent effect on lung cancer risks.

In addition to this, a meta analysis of 13 randomized trials, including almost 50,000 subjects, reported the total number of lung cancers incurred in folic acid containing supplements was 272, compared to 253 in those not taking folic acid (54). This represents a non-significant 8 percent increase in lung cancer incidence.

Interestingly enough, a small trial on 73 smokers found that a folate and vitamin B12 supplement reduced bronchial metaplasia (53). Although difficult to extrapolate this finding to lung cancer prevention, it does provide fairly weak evidence that, given folic acid’s apparent negative (or at least non positive) effect on pulmonary neoplasms, vitamin B12 may help prevent lung cancer.

If vitamin B12 is beneficial to lung cancer, it might explain why the WAFACS and SEARCH trials observed no effect despite folic acid: the folic acid and B12 cancelled each other out. Unfortunately, trials on vitamin B12 alone have not been conducted and thus we have little data to suggest what kind of effect it would have on this type of cancer.

A combination of two other B-vitamins, riboflavin and niacin, were tested together for their effect on lung cancer in a study mentioned earlier (8). This trials found that subjects taking these vitamins experienced a non-signficant 16% increase in lung cancer compared to those who did not. Unfortunately, this trial makes it difficult to say whether this effect might have been due to the riboflavin, niacin, or statistical chance.

The results of the WHO’s Coronary Drug Project may help explain this finding (56). The study tested niacin alone and found an apparent, though non-significant increase in lung cancer in the roughly 1,100 men taking it instead of a placebo.

These previous 2 studies do not clearly demonstrate a danger of niacin, simply suggest one weakly.

Conclusion: Effect of Nutritional Factors on Lung Cancer Risk

The following is my summary of the effects of various nutrients on lung cancer, as well as the strength of the evidence:

B-carotene: ↑ risk in smokers and the asbestos exposed***   ↓ risk in non-smokers*
Retinol: ↓ risk*
Zinc: ↓ risk*
Selenium: ↓ risk when correcting selenium deficiency**
Linoleic acid (omega-6): ↑ risk***
Vitamin D: ↓ risk*
Vitamin E (a-tocopherol): ↓ risk up to 400 IU’s*
↑ risk at 600 IU’s*
Folic acid: ↑ risk**
Vitamin B6: No apparent benefit or harm*
Vitamin B12: Insufficient evidence to suggest benefit or harm*
Niacin: ↑ risk*

* – Weak evidence
** – Decent evidence
*** – Moderate evidence
**** – Strong evidence

With a lot of weak evidence, the most well supported dietary advice for avoiding lung cancer might be as follows:

  • Avoid linoleic acid, found in vegetable oils, seeds, nuts, and, depending on how they were raised, fat from chickens and pigs.
  • If you smoke, avoid B-carotene, found in some multivitamins, carrots, and sweet potatoes (the orange ones). B-carotene absorption is also increased by cooking and in the presence of dietary fat.
  • Avoid folic acid, primarily the synthetic variety (arguments about the effect of natural source folate rages on), found in some multivitamin supplements and refined grain (e.g. white bread and some white rice).
  • Obtain adequate (but not excessive) selenium, which can be found in fish (including shellfish), meat, grains, mushrooms (particularly crimini), and seeds (which are usually rich in linoleic acid however).




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s