expert opinion

Folic Acid and the Risk of Lung Cancer – A Response to a Scientific Publication

November 30, 2009


Prof. Dr. med. vet. K. Pietrzik, Institute for Nutrition and Food Sciences Dept. of Nutritional Pathophysiology, Rheinische Friedrich-Wilhelms-University Bonn, Germany


“The publication under discussion (1) involves the combined evaluation of 2 placebo-controlled, double-blind, randomized clinical studies (RCTs): NORVIT, Norwegian Vitamin Trial (2), and Western Norway B Vitamin Intervention Trial (3). Both studies had the objective of lowering the homocysteine level in patients with existing cardiovascular diseases so as to consequently lower the risk of cardiovascular diseases. Different vitamin combinations were used for this in each multi-section study (vitamin B9 0.8 mg/day plus vitamin B12 0.4 mg/day, plus 40 mg vitamin B6 /day, and folic acid together with B12 or Vitamin B6 alone).

The NORVIT study attracted major media attention immediately after the initial presentation of the results during the cardiology congress on September 6, 2005 in Stockholm, when it was reported that in each of the vitamin B9 (folic acid) groups (on the one hand with B12 and B6, on the other with B12 alone) 33 cancer cases rose while only 25 cases were observed in the placebo group (vitamin B6 section 22 cases). Although the differences were not statistically significant (p=0.08) they attracted media attention when they were given as 40%. Since then all corresponding studies have been examining whether a similar connection can be found.

When the final results of the NORVIT study were then published (2) the following year after careful evaluation (including histological investigation), the media interest was comparatively slight since there was no further mention of an increased cancer risk. The publication makes it clear that 40 and 39 cancer cases (with and without vitamin B6) were diagnosed in the folic acid sections, while the placebo group also had 40 cases, so that there was no higher rate.

Because of the previous negative reporting on the NORVIT study, the WENBIT study was then discontinued and the participants were collated without further treatment as part of a follow-up observation (using the Norwegian register of deaths and the causes of death noted in it) of the frequency of cancer cases in the individual treatment groups. A comparable follow-up investigation was also undertaken with the participants of the NORVIT study.

The results now produced with the combined evaluation suggest the conclusion that administration of folic acid has significantly raised the cancer risk in particular for lung cancer. After an average of 39 months of treatment and an additionally observed 30-month follow-up period after the study, cancer was diagnosed (HR 1.21; 95% CI, 1.03-1.41; p=.02) in 341 participants (10%) who received folic acid plus vitamin B12, as opposed to 288 participants (8.4%) who were not given treatment. A total of 136 (4%) of those who received folic acid plus vitamin B12 died of cancer as opposed to 100 (2.9%) of those untreated (HR, 1.38; 95% CI, 1.07-1.79; p=.01). A total of 548 patients (16.1%) who received folic acid plus vitamin B12 died from other causes as opposed to 473 (13.8%) of those untreated (HR, 1.18; 95% CI, 1.04-1.33; p=0.1).The results were due mainly to an increase in the incidence of lung cancer in the treatment group. B6 treatment was not significantly associated with any effects.

A critical examination of the data makes it clear that a moderate improvement in folate status due to the study was evidently associated with a reduction in cancer risk, since the authors find the lowest cancer risk in the study participants whose serum folate (and vitamin B12) values lie in the 2nd quarter (serum folate values between 3.81 and 10.56µg/ml, serum B12 values between 456.4 and 595.9 pg/ml).They therefore take this 2nd quarter as reference for the assessment of the risk increase in the 4th quarter. Not mentioned is whether the study participants differ significantly in the lowest quarter (lowest folate status) from those in the 2nd quarter in relation to cancer frequency. Likewise unreported is whether significant results also arise in a comparison between the 1st and 4th quarter. Should this not be the case (which is to be assumed on the basis of the evaluation carried out), the treatment would accordingly also show no statistical effect, i.e. the improvement in the 2nd quarter would be compensated for by the deterioration in the 4th quarter. The considerations presented here do not exclude the possibility that administration of high doses of folic acid (and vitamin B12) is nevertheless followed by an increased cancer risk; however targeted investigations are required for this which permits a monocausal attribution.

Thus for example at the same time as the administration of folic acid (fourfold RDA 200µg folic acid corresponds to 400µg food folate) a more than hundredfold overdosage of vitamin B12 (0.4 mg) was instituted (RDA 3 µg). Since both vitamins are involved in DNA synthesis (this is also discussed in the publication, among other things, as a cause of the vitamin-determined increase in the cancer risk) a clear causal attribution cannot be made. The authors do conjecture that the deciding cause is attributed to folic acid, since a serum concentration dependence between the named quartiles is observed which however is not the case for vitamin B12. This conclusion is however not convincing inasmuch as after more than hundredfold overdosage with vitamin B12 (in spite of limited resorption), over the years (of treatment) in all groups such a high steady state may have arisen that a serum concentration dependence is no longer discernible. This is not to say that the author of this response attributes a causal role to vitamin B12 in the cancer events described here, only that the argumentation in the presented study must be comprehensible, which is not the case here.

The authors further discuss as possible causes for the increase in cancer risk the fact that after administration of folic acid in high doses, non-metabolized folic acid also increasingly occurs in the blood, which purely in theory might influence the pathogenesis of the cancer events, since later studies have shown that unmetabolized folic acid reduces the cytotoxicity of natural killer cells, so that the immune defenses are disturbed by cancer cells (4).Unmetabolized folic acid in the serum is however only observed regularly when the daily dosage of folic acid exceeds 400µg (the studies each applied 800µg).


There is therefore no reason to modify the present recommendations for the prevention of neural tube defects with 400µg/day. Likewise the recommendation to optimize the folate status must be upheld, so as to reduce the risk of cancer (as evidently also shown in the studies for the group with the lowest risk), cardiovascular diseases (see USA and HOPE as well as VISP study) and others associated with its deficiency as well.

In contrast to folic acid, the discussion does not relate to natural folates as they occur in foodstuffs. The more of this is supplied, the less the risk not only of neural tube defects but also of cancer, cardiovascular diseases and other diseases. If the vitamin is supplied in its synthetic form however, there is no reason to change current practice in the common dosage area (prevention of neural tube defects and in food enrichment) because of the results presented here.

A similar risk assessment is also reached by the authors of an editorial on the presented publication which was published in the same issue (5). According to this, the legally compulsory folic acid enrichment of basic foodstuffs prescribed since 1.1.1998 and the widespread ingestion of (multi)vitamin preparations should have led to an influence on the cancer rate (as described in the study) of alarming proportions, which is however not the case. When one further looks at the change in the levels of folate concentration in the blood after the enrichment in the USA, an increase in the lung cancer rate in particular should have occurred (according to Ebbing et al. 2009).

On the contrary, since 2001 a significant drop not only in the cancer rate generally, but especially also in the lung cancer rate in men and women has been described in the USA. The editorial in particular refers to the positive long-term effects of folic acid, whereby the USA data clearly show that the ingestion of folic acid is inversely associated with the cancer rate. In this connection it cites a study by Wolpin et al. in 2008 (6) in which plasma level measurements existed for folate before cancer was diagnosed. Here too a significant relationship with colorectal cancer as well as with overall mortality both before and after the enrichment of basic foodstuffs is evident. The reduction in cancer mortality was up to 50 percent.

Bonn, November 2009


  1. M. Ebbing et al. Cancer Incidence and Mortality after treatment with folic acid and Vitamin B12. JAMA. 2009; 302(19):2119–2126.
  2. Bønaa KH, Njolstad I, Ueland PM, et al; NORVIT Trial Investigators. Homocysteine lowering and cardiovascular events after acute myocardial infarction. N Engl J Med. 2006; 354(15):1578–1588.
  3. Ebbing M, Bleie O, Ueland PM, et al. Mortality and cardiovascular events in patients treated with homocysteine-lowering B vitamins after coronary angiography: a randomized controlled trial. JAMA. 2008; 300(7):795–804.
  4. Troen AM, Mitchell B, Sorensen B, et al. Unmetabolized folic acid in plasma is associated with reducednatural killer cell cytotoxicity among postmenopausal women. J Nutr. 2006; 136(1):189–194.
  5. Bettina F. Drake; Graham A. Colditz. Assessing Cancer Prevention Studies - A Matter of Time. JAMA. 2009; 302(19):2152–2153.
  6. Wolpin BM, Wei EK, Ng K, et al. Prediagnostic plasma folate and the risk of death in patients with colorectal cancer. J Clin Oncol. 2008; 26(19):3222–3228.