Despite potential, data on vitamin D’s benefits for cancer remain inconclusive

Researchers have long sought to determine whether vitamin D has a true protective effect against cancer.

Laboratory-based studies have shown vitamin D — a fat-soluble vitamin that promotes bone health —encourages cell differentiation, affects blood vessel formation, is anti-inflammatory and boosts the immune system.

“All of these different pathways are relevant for cancer development,” Kimmie Ng, MD, MPH, director of clinical research at Dana-Farber Cancer Institute’s Gastrointestinal Cancer Center, told HemOnc Today. “Because of these laboratory data, there is interest in vitamin D as a potential chemopreventive agent.”

Source: Dana-Farber Cancer Institute.

This interest is not limited to one malignancy.

Marjorie L. McCullough, ScD, RD, epidemiologist at American Cancer Society, and colleagues found that individuals with vitamin D deficiency had a 31% higher risk for colorectal cancer.

In another study, McDonnell and colleagues showed women with vitamin D levels higher than 60 ng/mL had an 80% lower risk for breast cancer than those who had levels less than 20 ng/mL.

Still, the nature of these analyses — which often are retrospective and population based — could introduce bias.

“We need to wait for a rigorously designed confirmatory randomized trial before we start recommending vitamin D for prevention and treatment of various cancers,” Ng said. “However, many clinicians are checking plasma vitamin D levels in the blood and at least trying to increase levels to 20 ng/mL or 30 ng/mL because it is also beneficial for bone health. Although the data on whether it’s helpful and protective for cancer is not as fully established, there are no harmful side effects from getting people up into the acceptable range.”

HemOnc Today spoke with oncologists and epidemiologists about data that have shown vitamin D may help prevent cancer or improve outcomes after diagnosis, the need to establish universal levels of adequate vitamin D, why associations between vitamin D and cancer outcomes may be confounded by other patient-level variables, and the likelihood that a randomized controlled trial could generate definitive conclusions.

‘No optimal level’

Vitamin D is a naturally occurring vitamin that has functions needed for human health.

“It’s something we need as part of our metabolism,” Stephen M. Ansell, MD, PhD, chair of the lymphoma group at Mayo Clinic in Rochester, Minnesota, and a HemOnc Today Editorial Board Member, said in an interview. “The predominant need for it is to ensure bone health and bone integrity. We’ve learned in recent years it has many other roles in cell metabolism and growth.”

Vitamin D helps the body absorb and maintain adequate serum calcium and phosphate concentrations to enable normal mineralization of bone.

“Vitamin D is different because it can increase calcium absorption in the intestine, and it is important to maintain calcium levels in the human body,” Song Yao, PhD, associate professor of oncology in the department of cancer prevention and control at Roswell Park Comprehensive Cancer Center, said in an interview.

Sun exposure, supplements and some foods provide vitamin D.

“Even though there is vitamin D fortification of milk and other food items, it is very hard to get a big enough dose of vitamin D through dietary sources to achieve the levels we think are potentially protective,” Ng said. “Sun exposure is very effective, but it is not the safest way to boost vitamin D levels.”

A chemopreventive level of vitamin D remains to be defined.

“There is not a clear optimal blood level,” Brian M. Wolpin, MD, MPH, director of Gastrointestinal Cancer Center and Hale Family Research Center and oncologist at Dana-Farber Cancer Institute and associate professor of medicine at Harvard Medical School, told HemOnc Today. “For cancer-related outcomes, it is not entirely clear what the optimal level should be.”

The nonprofit Vitamin D Council recommends adults maintain serum levels of 50 ng/ml, with deficiency defined as 0 ng/mL to 40 ng/mL and toxic levels defined as greater than 150 ng/mL.

“There is a lot of controversy about what defines a sufficient vitamin D level,” Ng said. “The data aren’t sufficient to recommend higher levels to affect the clinical endpoints of cancer or heart disease. The blood level of 25 hydroxyvitamin D — the main circulating form — should be 20 ng/mL to maintain bone health. Endocrinologists and those who research vitamin D as an anticancer or chemopreventive agent feel a higher level of 30 ng/mL or above is preferred.”

For adequate bone health, the National Academy of Medicine recommends intake of 600 IU per day among adults and 800 IU for those aged older than 70 years.

A fear of vitamin D deficiency has led to a trend of increased supplementation.

Rooney and colleagues used national survey data to determine 3% of the population exceeded the tolerable upper limit of 4,000 IU daily, putting them at risk for adverse effects. Eighteen percent of the population reported 1,000 IU daily, suggesting intentional vitamin D supplementation.

“It is very hard to get too much, but guidelines currently recommend 4,000 IU as the tolerable upper intake level,” Ng said. “People get a lot higher doses without any significant side effects. A blood level of 100 ng/mL would start to correlate with toxicity, but it’s almost impossible to achieve that level with how much vitamin D deficiency there is.”

Too much vitamin D can increase the amount of calcium in blood and urine, thereby increasing risk for kidney stones. There also is a chance of interaction with anticancer therapy.

“There is not a substantial risk that you will compromise chemotherapy, but one should still talk with their treating physician,” Ansell said.

Vitamin D deficiency is more prevalent.

“It is unclear whether it is due to increased indoor activities, use of sunscreen, high levels of obesity or all of the above, but there is an increasing epidemic of vitamin D deficiency,” Ng said.

Denise Rokitka, MD, PhD, assistant professor in the department of pediatric oncology and director of Roswell Park Comprehensive Cancer Center’s long-term follow-up clinic, and colleagues determined that identifying vitamin D levels among childhood cancer survivors may help prevent secondary cancers and chronic disease.

Their study included 139 adult survivors of childhood cancers. Results showed 34% were vitamin D deficient, 39% had insufficient levels and 27% had sufficient levels.

Further, 68.3% of survivors who reported supplementation still had insufficient or deficient levels. Survivors aged older than 30 years and those who were overweight appeared most likely to have vitamin D insufficiency or deficiency.

“The long-term effects of therapy and quality of life among adult survivors of childhood cancers have become an area of intense investigation,” Rokitka said in a press release. “My hope is that future studies will help clarify vitamin D’s role in prevention and health maintenance for these survivors and determine the extent to which vitamin D supplements can improve outcomes.”

Protective effects

Although the benefits of vitamin D on bone diseases are well known, there is growing evidence that adequate levels may reduce cancer risk.

“In more recent years, vitamin D has shown more potential in cancer prevention, which is why we are studying it,” Yao said.

Budhathoki and colleagues conducted a prospective study in Japan that included 3,301 incident cases of cancer and 4,044 randomly selected participants. Researchers divided participants into quartiles based on sex and season-specific distribution of 25-hydroxyvitamin D, or 25(OH)D, for which the first quartile had the lowest vitamin D levels and the fourth had the highest.

Vitamin D plasma concentration levels in quartile two (adjusted HR [aHR] = 0.81; 95% CI, 0.7-0.94), three (aHR = 0.75; 95% CI, 0.65-0.87) and four (aHR = 0.78; 95% CI, 0.67-0.91) reduced risk for total cancer compared with the first quartile. The strongest association occurred for liver cancer (fourth quartile, aHR = 0.45; 95% CI, 0.26-0.79).

Consistent associations of vitamin D’s protective effects have been observed for colorectal cancer.

The vitamin D receptor is present in many tissues throughout the body, including colorectal mucosa. The active form of vitamin D binds to the receptor, which — in turn — influences expression of many genes involved in cell cycle regulation, McCullough said.

“Vitamin D supplementation has been shown to modulate expression of the adenomatous polyposis coli gene, [which is] important in colorectal carcinogenesis,” McCullough told HemOnc Today. “Other mechanisms specific to the large bowel may include promoting detoxification of DNA-damaging lithocholic acid and improving gut mucosal integrity and immunity.”

In a study published in Journal of the National Cancer Institute, McCullough and colleagues pooled data from 17 cohorts that included 5,706 patients with colorectal cancer and 7,107 controls.

Researchers determined prediagnostic vitamin D levels less than 30 nmol/L (12 ng/mL) were associated with a 31% increased risk for colorectal cancer (RR = 1.31; 95% CI, 1.05-1.62) compared with levels 50 nmol/L to 62.5 nmol/L (15.7 ng/mL to 19.5 ng/mL).

Further, vitamin D levels above sufficiency decreased colorectal cancer risk. For example, levels of 87.5 nmol/L up to 100 nmol/L (35 ng/mL-40 ng/mL) were associated with a 27% reduced risk (RR = 0.73; 95% CI, 0.59-0.91).

“Health agencies do not recommend vitamin D for the prevention of colorectal cancer,” McCullough said in a press release. “This study adds new information that agencies can use when reviewing evidence for vitamin D guidance and suggests that the concentrations recommended for bone health may be lower than would be optimal for colorectal cancer prevention.”

In an analysis of patients from two pooled randomized trials by McDonnell and colleagues, researchers found women with concentrations of 60 ng/mL or higher had an 80% lower risk for breast cancer than those with concentrations less than 20 ng/ml in adjusted models (HR = 0.2; P = .03).

Because the study was to women with postmenopausal breast cancer, more research is needed on premenopausal breast cancer.

“Nonetheless, this paper reports the strongest association yet between serum vitamin D and reduction in risk of breast cancer,” study author Cedric F. Garland, DrPH, adjunct professor in the department of family medicine and public health at UC San Diego, said in a press release.

Effects on cancer outcomes

Research also is underway to determine whether vitamin D supplementation can improve outcomes among individuals with cancer.

Preclinical research showed modified vitamin D may break barriers to tumors, enabling therapy to reach cancer-isolated cells.

Sherman and colleagues investigated vitamin D and its effect on pancreatic cancer in mouse models. They found the combination of modified vitamin D and chemotherapeutics led to reduced tumor volume and a 57% increase in survival compared with chemotherapy alone.

“The vitamin D was binding to supportive cells in the tumor, which led to reduced inflammatory signals,” Wolpin said about the study’s results.

This improved penetration of chemotherapy into the tumor.

“Based on those findings, several clinical trials in pancreatic cancer are testing vitamin D analogs to see if they will lead to better outcomes for patients,” Wolpin said.

A clinical trial at Dana-Farber is underway to test standard chemotherapy with or without a vitamin D analog among patients with advanced pancreatic cancer.

In another study, Wolpin and colleagues analyzed five U.S. prospective cohorts of 493 patients with pancreatic cancer. Researchers found prediagnostic plasma 25(OH)D levels of 30 ng/mL or higher appeared associated with longer pancreatic cancer survival.

The mean 25(OH)D level prior to diagnosis was 24.6 ng/mL, and 165 (33%) patients were considered vitamin D insufficient.

In adjusted analyses, patients with sufficient 25(OH)D levels demonstrated reduced risk for death compared with patients deficient in vitamin D (HR = 0.66; 95% CI, 0.49-0.9).

“When considering these findings together with previously reported preclinical data in pancreatic cancer models, agonists of the vitamin D receptor are a potentially attractive therapeutic approach for investigation in this highly lethal malignancy,” Wolpin and colleagues wrote.

Preclinical data also led to the phase 2 SUNSHINE trial, in which Ng and colleagues compared the use of low-dose (400 IU daily) vs. high-dose (8,000 IU loading dose for 2 weeks; 4,000 IU daily thereafter) vitamin D among patients with metastatic colorectal cancer.

“We started to do a lot of observational and epidemiology studies where we looked at healthy individuals first, and we noticed people who had highest vitamin D had significantly reduced risk for developing colon cancer,” Ng said. “We then looked at patients with all stages of colon cancer and found that higher levels of vitamin D in the blood were associated with significantly better survival from colon cancer.”

Results showed patients who received high-dose vitamin D experienced longer median PFS than those who received the low dose (12.4 months vs. 10.7 months; HR = 0.66; 95% CI, 0.45-0.99). They also exhibited a better disease control rate (100% vs. 94%; P = .05).

“That was the first randomized trial of vitamin D as a potential treatment for metastatic colorectal cancer and it showed very promising results,” Ng said, adding that the group is designing a confirmatory phase 3 study with NCI.

Yao, Lawrence H. Kushi, ScD, director of scientific policy at Kaiser Permanente Northern California Division of Research, and colleagues also have been evaluating vitamin D for breast cancer.

In a prospective cohort study, researchers collected data from 1,666 women (mean age, 58.7 years) from Kaiser Permanente Northern California to determine how serum 25(OH)D levels collected within 2 months of breast cancer diagnosis correlated with prognostic characteristics and outcomes.

At baseline, 48% of the patient population was vitamin D deficient, and 35% had vitamin D insufficiency. Researchers observed lower 25(OH)D concentrations among black and Hispanic women, current smokers and younger women.

Further, mean serum 25(OH)D concentrations were lower among women with advanced-staged tumors, and concentrations were lowest among premenopausal women with triple-negative cancer (luminal A, 20 ng/mL; luminal B, 19.8 ng/mL; HER2 enriched, 19.3 ng/mL; triple negative, 18.7 ng/mL).

Higher 25(OH)D levels appeared associated with improved OS (HR = 0.54; 95% CI, 0.4-0.72) and breast cancer-specific survival (HR = 0.58; 95% CI, 0.38-0.9).

“We did see higher vitamin D levels at the time of cancer diagnosis were associated with better breast cancer survival, which is consistent with some other population-based studies,” Yao said. “There are other studies [that did not show] such findings, so these kinds of studies can be complicated. More research is certainly needed.”

‘Therapeutic opportunity’

There has been some study into the role of vitamin D in hematologic malignancies.

Kelly and colleagues evaluated data from two cohorts of previously untreated patients with newly diagnosed follicular lymphoma from SWOG (n = 183) clinical trials or the Lymphoma Study Association (LYSA) PRIMA trial (n = 240).

Vitamin D deficiency was defined as levels less than 20 ng/mL in the SWOG cohort and less than 10 ng/mL in the LYSA cohort. Using these definitions, 15% of the SWOG cohort was vitamin D deficient (median, 31 ng/mL) and 25% of the LYSA cohort was considered deficient (median, 17 ng/mL).

After median follow-up of 5.4 years in the SWOG cohort, researchers observed no association between vitamin D deficiency and cause of death or clinical response (OR for complete response = 1.14; 95% CI, 0.46-2.81). However, patients who were vitamin D deficient showed significantly shorter PFS (HR = 2; P = .011) and OS (HR = 3.57; P = .003).

After median follow-up of 6.6 years in the LYSA cohort, an equal proportion of patients who were and were not deficient demonstrated clinical response. Although patients with vitamin D deficiency showed inferior PFS (HR = 1.66; P = .013), the association between lower vitamin D and OS did not reach statistical significance (HR = 1.84).

The finding that levels of vitamin D correlated with follicular lymphoma outcomes is “not surprising,” Ansell told HemOnc Today when the study was published.

“Other studies that included patients with diffuse large B-cell lymphoma and chronic lymphocytic leukemia have also shown pretreatment vitamin D levels to be prognostic,” he said. “This study now confirms that vitamin D levels are also important in follicular lymphoma.”

Several mechanisms may be driving this association.

“Low vitamin D levels may interfere with macrophage function, and this may account in some part for the poor prognosis of patients with low vitamin D levels,” he said. “Macrophages play a significant role in the body’s response to monoclonal antibody therapy, and compromised macrophage function due to low vitamin D levels could make the response to therapy less effective.”

Additional studies have shown low vitamin D is correlated with poorer outcomes despite the use of standard therapies.

“A significant association between vitamin D levels and outcome has been observed in patients with aggressive lymphomas, and this may suggest an association with the biology of the tumor. The lymphoma may be using vitamin D to promote growth and progression, and so causing the levels to be suppressed,” Ansell said. “Aggressive cancer may dysregulate vitamin D metabolism and thereby make low levels more likely to be associated with a bad outcome.”

However, adequate vitamin D levels also may be a surrogate marker for a healthy lifestyle, regardless of the cancer population under study.

“The fact that serum vitamin D levels [in the study by Kelly and colleagues] were associated with performance status — and also with adverse prognostic factors, such as anemia — suggests patients who are sick may have lower vitamin D levels, and the poor prognosis is related to being ill, not just the vitamin D level,” Ansell said.

“The key question, however, is whether the low vitamin D levels are in some way a surrogate for an underlying component of tumor biology or whether this constitutes a therapeutic opportunity,” he added. “Alternatively, it could be that the tumor cells themselves account for the low vitamin D levels and therapy will need to be directed at the tumor cell itself.”

Still, the results of the analysis by Kelly and colleagues should be interpreted with caution, Ansell said.

Because not all patients had their serum vitamin D levels measured, there is a potential for selection bias. Further, the different levels used for vitamin D sufficiency in the two cohorts may have introduced biases.

“The biological relevance of a low vitamin D level is an area that needs investigation,” Ansell said.

Limitations and confounding factors

The potential biases of the study by Kelly and colleagues reflect a larger trend in this area: A lack of consensus on dosing and measurements can complicate research into vitamin D.

“One thing that challenges research in this field is different laboratories using different methods to measure vitamin D,” McCulloughsaid. “Studies conducted have shown to have widely varying measurements of serum vitamin D so, depending on what they use in the laboratories, the absolute concentrations can vary quite a bit.”

Many studies investigating the association between vitamin D and cancer risk are observational and epidemiological, so they reveal associations but do not prove cause and effect. Population-based studies also open the door for confounding factors.

“It’s hard to say that only vitamin D level explains patient outcomes,” McCullough said.

Experimental studies have shown a variety of anticancer properties of vitamin D antiproliferation, Yao said.

“The preclinical literature is very solid,” he said. “Other studies have been less conclusive.”

Because risk factors vary by cancer type, vitamin D’s effect also is likely to vary.

“To date, the most consistent association of higher levels of vitamin D with lower cancer risk has been for colorectal cancer,” McCullough said. “It is more complicated for other cancers, like prostate cancer. There may be a lower risk for more aggressive forms with higher levels of vitamin D, but the data haven’t been consistent.”

Case-control studies for breast cancer have shown inverse associations; however, this could be related to changes in vitamin D from having breast cancer, whether due to treatments, the cancer itself or lifestyle changes since diagnosis, McCullough said.

“So, the majority of prospective studies have not shown lower breast cancer risk with higher blood vitamin D,” she said.

Vitamin D levels are influenced by diet, genetic makeup, climate and other characteristics.

“These factors all complicate exactly how to interpret things,” Ansell said. “These have to be put into multifactorial analysis.”

Obesity — often a poor prognostic marker for patients with cancer — is inversely related to vitamin D because it is fat soluble.

“In patients who are overweight or obese, their vitamin D is lower, which creates bias,” Yao said. “It’s quite hard to conduct a study where you can properly control all confounding factors.”

Further, sick patients may not absorb as much vitamin D.

“Vitamin D is associated with decreased cancer risk and better outcome if you have adequate levels,” Ansell said. “It’s sometimes complicated because sick patients might be indoors a lot or bedridden, which would decrease serum vitamin D, creating more difficult interpretation. We have to think, is it because they are sick or is the vitamin D associated with outcome?”

Yao agreed sun exposure can be a proxy for performance status.

“In that sense, a lot of the vitamin D research has a sense of confounding bias that we need to consider,” he said.

Genetic makeup is another factor.

“African-Americans are at higher risk for vitamin D deficiency because their skin has less ability to synthesize vitamin D from the sun,” McCullough said. “Although African-Americans can have lower vitamin D levels, their bone density isn’t worse, so some have even suggested their genetic background has a lower threshold of need for vitamin D.”

However, because black individuals often have higher incidence and mortality of certain cancers, how vitamin D may affect this is difficult to determine, McCullough added.

“We do think people of different backgrounds, ethnicities and geographies respond to vitamin D in different ways,” Ng said.

“African-Americans have significantly lower vitamin D in blood than whites, so it is possible that might contribute to the racial disparities in colon cancer incidence and mortality,” Ng added. “It’s difficult to do a study that includes enough patients with all ethnicities, but it is definitely something we are interested in exploring more.”

More research needed

Although there are a number of diseases in which vitamin D seems to be useful, the lack of confirmatory data leaves researchers wondering whether the association is only a correlation or due to a cause-and-effect relationship, Wolpin said.

“Most of the studies aren’t randomized, so there could be other reasons why they have lower cancer risk or higher survival that are not directly due to higher vitamin D,” he said.

Randomized trials stand to provide more robust information.

“The one weakness is that there haven’t been many randomized studies that looked at replacement vs. nonreplacement and showed there is a difference,” Ansell said.

The ongoing randomized VITamin D and OmegA-3 TriaL, or VITAL, includes 25,874 U.S. men and women.

Researchers are investigating whether taking daily dietary supplements of vitamin D3 at 2,000 IU or 1 g omega-3 fatty acids reduces risk for cancer, heart disease and stroke among people who do not have a prior history of these illnesses.

Another aim of the study is to determine whether supplementation can reduce incident total fractures and incident hip and nonvertebral fractures.

Randomized trials also can help identify a maximum tolerated dose of serum vitamin D that would be protective against cancer.

“We commonly use 20 ng/mL as a cutoff of vitamin D deficiency, but there is not a clear consensus on what level is enough for improved cancer-related outcomes,” Wolpin said. “To reach that answer, randomized clinical trials are needed.”

Until more trials are conducted, the benefit of vitamin D for cancer will remain controversial.

“The randomized controlled trial is the most definitive kind of study we can do to answer this question,” Wolpin said. – by Melinda Stevens

Click here to read the , “Should clinicians recommend vitamin D supplementation to patients with cancer to improve their outcomes?”

References:

Budhathoki S, et al. BMJ. 2018;doi:10.1136/bmj.k671.

Kelly JL, et al. J Clin Oncol. 2015;doi:10.1200/JCO.2014.57.5092.

McCullough MJ, et al. J Natl Cancer Inst. 2018;doi:10.1093/jnci/djy087.

McDonnell SL, et al. PLoS One. 2018;doi:10.1371/journal.pone.0199265.

Ng K, et al. J Clin Oncol. 2017;doi:10.1200/JCO.2017.35.15_suppl.3506.

Rokitka DA, et al. Int J Cancer Ther Oncol. 2016;doi:10.14319/ijcto.43.10.

Rooney MR, et al. JAMA. 2017;doi:10.1001/jama.2017.4392.

Sherman MH, et al. Cell. 2014;doi:10.1016/j.cell.2014.08.007.

Vitamin D Council. Position statement on supplementation, blood levels and sun exposure. Available at: www.vitamindcouncil.org/for-health-professionals-position-statement-on-supplementation-blood-levels-and-sun-exposure. Accessed Sept. 18, 2018.

Yao S, et al. JAMA Oncol. 2016;doi:10.1001/jamaoncol.2016.4188.

Yuan C, et al. J Clin Oncol. 2016;doi:10.1200/JCO2015.66.3005.

For more information:

Stephen M. Ansell, MD, PhD, can be reached at ansell.stephen@mayo.edu.

Marjorie L. McCullough, ScD, RD, can be reached at marji.mccullough@cancer.org.

Kimmie Ng, MD, MPH, can be reached at kimmie_ng@dfci.harvard.edu.

Brian M. Wolpin, MD, MPH, can be reached at brian_wolpin@dfci.harvard.edu.

Song Yao, PhD, can be reached at song.yao@roswellpark.org.

Disclosures: Wolpin reports research funding from Celgene. Ansell, McCullough, Ng and Yao report no relevant financial disclosures.