Feature

Data provide insight into 'not trivial' link between height, cancer risk

Photo of Leonard Nunney
Leonard Nunney

Taller individuals may be at higher risk for certain cancer types, according to study findings published in Proceedings of the Royal Society B.

Previous research has shown factors early in life — such as nutrition, health and social conditions — independently influence height and cancer risk.

“I tested the alternative hypothesis that height increases cell number and that having more cells directly increases cancer risk,” Leonard Nunney, PhD, professor of biology at University of California, Riverside, said in a press release.

Nunney analyzed data from four large-scale surveillance projects including 23 cancer categories. All of the studies showed that height is associated with cancer risk. The overall cancer risk increased by approximately 10% per 4-inch increase in height.

“Tall individuals are at increased risk for almost all cancers,” Nunney said in the press release. “Skin cancers, such as melanoma, showed an unexpectedly strong relationship to height, which may be due to the higher levels of the hormone IGF-1 in taller adults.”

HemOnc Today spoke with Nunney about his study and what future research on this topic should entail.

Question: What prompted this research?

Answer: I am an evolutionary biologist and my interest is in applying evolutionary ideas to problems in cancer. My focus tends to compare different species of animals. Specifically, I am interested in how large-sized, long-lived animals have prevented cancer. The reason for this focus comes from the basic model of how cancer arises — multistage carcinogenesis — where mutations are either inherited or accumulated in cells due to somatic mutations in life, and some combination of these mutational changes will initiate a cancer. This basic model makes clear predictions. One of the strongest predictions is that, if all else is equal, a small animal — such as a mouse — with a short life should get a lot less cancer than a larger, long-lived animal, such as a human. However, they do not. Large, long-lived animals have evolved additional ways of preventing cancer. This argument does not apply to variation in size within a species. If the model is correct, we would still expect that within a species, bigger individuals should get more cancer than smaller individuals.

Q: What has previous research on this shown?

A: When I started research on this topic, there were no clear data on whether there was a height-cancer link in humans, with height serving as a proxy for lean body size. We now have clear evidence that taller individuals are more susceptible to cancer than shorter individuals. This association has been significantly established by large-scale cancer studies, the largest of them being the Million Women study. This and other research has demonstrated that not only was there an increased overall risk for cancer among taller individuals, but that this risk extended across a range of different cancer types. This is what the model predicts — a taller individual has more cells in all tissues, so that in any given tissue, they have a higher risk for accumulating the critical set of mutations that lead to cancer.

An alternative explanation was that the height effect was not associated with the number of cells, but instead related to events occurring early in life that not only resulted in making individuals taller, but also caused cells to be more prone to cancer. Levels of growth hormone, such as IGF-1, were considered possible mediators of this effect.

Q: How did you conduct the current study?

A: I wanted to see if the effect we see in humans could be most parsimoniously explained by changes in cell number with height. For this study, I exploited studies that were already published — the biggest of them the Million Women study in which more than 1 million women were tracked for almost 10 years. My study was a meta-analysis of four data sets. Height needed to be translated into cell number in order to test the model. I used values close to the BMI standard, which assumes that body mass scales as height squared. The precise link to lean body mass is slightly less than a squared relationship for men and slightly more for women.

Q: Can you elaborate on the findings?

A: It turns out that the data fit closely with my numerical predictions based on cell number alone. Furthermore, the effects of height in increasing the risk for specific cancers was not correlated with the effect of BMI in increasing cancer risk, which is what I expected since being overweight does not increase cell number and likely has a very different causation. Another height-related observation is that men get more nonreproductive cancer than women, but only about one-third of that increased risk can be accounted for by the height difference.

Q: Are there certain cancer types that appear to be more associated with height?

A: A few cancers show little or no association with height. These cancers may have a significant environmental causation masking the height effect. For example, cervical cancer does not scale with height, but we know that almost all cervical cancer is linked to viral infection, such as HPV. Similarly, mouth and oropharyngeal cancer have the same pattern and they are linked to significant environmental factors, such as smoking.

Two types of cancer appeared to have a notably stronger association with height than expected, one of which was thyroid cancer. This cancer also was associated with a strong geographical component. The women in the Korean study in particular showed an incredibly strong relationship between thyroid cancer and height — a 54% increased risk per 10 cm in height — that was not apparent in other studies in the U.S. and Europe.

Even more interesting is the unusually strong association between height and skin cancer, notably melanoma. Why does melanoma scale more strongly with height than one would expect? One proposal that I have come up with is that, because tall individuals have higher levels of IGF-1, this may result in a slightly higher division rate of skin cells. Even a very small increase in the division rate with height could account for the observed pattern.

PAGE BREAK

Q: What are the clinical implications of the findings?

A: The primary implication for clinicians is that they should be aware that very tall individuals are at an increased risk for almost every cancer. Although the effect is modest, it is not trivial and should not be ignored.

Q: Should these findings be incorporated into clinical practice?

A: Given the stronger-than-expected link between height and incidence of skin cancer, clinicians should be extra vigilant in monitoring for signs of skin cancer in very tall individuals. For example, a 6-foot-1-inch woman has a 50% greater overall cancer risk than a woman who is 5 feet tall, and her risk for skin cancer is roughly doubled.

Q: What is next for research on this topic?

A: My work on human body size and cancer was driven by the need to confirm an important prediction of the multistage model. Another species in which this same effect is expected and appears to be strongly supported is the domestic dog: Larger dog breeds get more cancer than smaller-sized dogs. However, the more important future of this research is in the search to understand why the effect of size in increasing cancer is not seen when we compare a huge animal, such as an elephant, to a tiny animal, such as a mouse. Now that we can be confident that larger-sized individuals are more prone to cancer than smaller-sized individuals within a species, it is time to focus on comparative studies that involve species of different body size. What features of the biology of elephants and whales prevent body-size prediction from being true? We need to understand what adaptive changes have occurred in these large long-lived animals that suppresses their risk for cancer. If we can establish this, we may be able to use this information to develop therapeutic methods for minimizing cancer risk in humans.

Q: Is there anything else that you would like to mention?

A: Traditional cancer biologists have done incredibly wonderful things in advancing our understanding of cancer and of therapies to control it, but I think there remains an important place for the comparative evolutionary approach. This approach has never been exploited in the past, but natural selection has clearly found ways of suppressing cancer in the largest-sized mammals. With an evolutionary approach, we are asking what is different about these animals, and if can they show us new avenues for controlling cancer. – by Jennifer Southall

References:

Green J, et al. Int J Epidemiol. 2011;doi:10.1093/ije/dyy065.

Nunney L. Proc Biol Sci. 2018;doi:10.1098/rspb.2018.1743.

Sung J, et al. Am J Epidemiol. 2009;doi:10.1093/aje/kwp088.

For more information:

Leonard Nunney, PhD, can be reached at University of California, Riverside, 900 University Ave., Riverside, CA 92521; email: nunney@ucr.edu.

Disclosure: Nunney reports no relevant financial disclosures.

Photo of Leonard Nunney
Leonard Nunney

Taller individuals may be at higher risk for certain cancer types, according to study findings published in Proceedings of the Royal Society B.

Previous research has shown factors early in life — such as nutrition, health and social conditions — independently influence height and cancer risk.

“I tested the alternative hypothesis that height increases cell number and that having more cells directly increases cancer risk,” Leonard Nunney, PhD, professor of biology at University of California, Riverside, said in a press release.

Nunney analyzed data from four large-scale surveillance projects including 23 cancer categories. All of the studies showed that height is associated with cancer risk. The overall cancer risk increased by approximately 10% per 4-inch increase in height.

“Tall individuals are at increased risk for almost all cancers,” Nunney said in the press release. “Skin cancers, such as melanoma, showed an unexpectedly strong relationship to height, which may be due to the higher levels of the hormone IGF-1 in taller adults.”

HemOnc Today spoke with Nunney about his study and what future research on this topic should entail.

Question: What prompted this research?

Answer: I am an evolutionary biologist and my interest is in applying evolutionary ideas to problems in cancer. My focus tends to compare different species of animals. Specifically, I am interested in how large-sized, long-lived animals have prevented cancer. The reason for this focus comes from the basic model of how cancer arises — multistage carcinogenesis — where mutations are either inherited or accumulated in cells due to somatic mutations in life, and some combination of these mutational changes will initiate a cancer. This basic model makes clear predictions. One of the strongest predictions is that, if all else is equal, a small animal — such as a mouse — with a short life should get a lot less cancer than a larger, long-lived animal, such as a human. However, they do not. Large, long-lived animals have evolved additional ways of preventing cancer. This argument does not apply to variation in size within a species. If the model is correct, we would still expect that within a species, bigger individuals should get more cancer than smaller individuals.

Q: What has previous research on this shown?

A: When I started research on this topic, there were no clear data on whether there was a height-cancer link in humans, with height serving as a proxy for lean body size. We now have clear evidence that taller individuals are more susceptible to cancer than shorter individuals. This association has been significantly established by large-scale cancer studies, the largest of them being the Million Women study. This and other research has demonstrated that not only was there an increased overall risk for cancer among taller individuals, but that this risk extended across a range of different cancer types. This is what the model predicts — a taller individual has more cells in all tissues, so that in any given tissue, they have a higher risk for accumulating the critical set of mutations that lead to cancer.

PAGE BREAK

An alternative explanation was that the height effect was not associated with the number of cells, but instead related to events occurring early in life that not only resulted in making individuals taller, but also caused cells to be more prone to cancer. Levels of growth hormone, such as IGF-1, were considered possible mediators of this effect.

Q: How did you conduct the current study?

A: I wanted to see if the effect we see in humans could be most parsimoniously explained by changes in cell number with height. For this study, I exploited studies that were already published — the biggest of them the Million Women study in which more than 1 million women were tracked for almost 10 years. My study was a meta-analysis of four data sets. Height needed to be translated into cell number in order to test the model. I used values close to the BMI standard, which assumes that body mass scales as height squared. The precise link to lean body mass is slightly less than a squared relationship for men and slightly more for women.

Q: Can you elaborate on the findings?

A: It turns out that the data fit closely with my numerical predictions based on cell number alone. Furthermore, the effects of height in increasing the risk for specific cancers was not correlated with the effect of BMI in increasing cancer risk, which is what I expected since being overweight does not increase cell number and likely has a very different causation. Another height-related observation is that men get more nonreproductive cancer than women, but only about one-third of that increased risk can be accounted for by the height difference.

Q: Are there certain cancer types that appear to be more associated with height?

A: A few cancers show little or no association with height. These cancers may have a significant environmental causation masking the height effect. For example, cervical cancer does not scale with height, but we know that almost all cervical cancer is linked to viral infection, such as HPV. Similarly, mouth and oropharyngeal cancer have the same pattern and they are linked to significant environmental factors, such as smoking.

Two types of cancer appeared to have a notably stronger association with height than expected, one of which was thyroid cancer. This cancer also was associated with a strong geographical component. The women in the Korean study in particular showed an incredibly strong relationship between thyroid cancer and height — a 54% increased risk per 10 cm in height — that was not apparent in other studies in the U.S. and Europe.

Even more interesting is the unusually strong association between height and skin cancer, notably melanoma. Why does melanoma scale more strongly with height than one would expect? One proposal that I have come up with is that, because tall individuals have higher levels of IGF-1, this may result in a slightly higher division rate of skin cells. Even a very small increase in the division rate with height could account for the observed pattern.

PAGE BREAK

Q: What are the clinical implications of the findings?

A: The primary implication for clinicians is that they should be aware that very tall individuals are at an increased risk for almost every cancer. Although the effect is modest, it is not trivial and should not be ignored.

Q: Should these findings be incorporated into clinical practice?

A: Given the stronger-than-expected link between height and incidence of skin cancer, clinicians should be extra vigilant in monitoring for signs of skin cancer in very tall individuals. For example, a 6-foot-1-inch woman has a 50% greater overall cancer risk than a woman who is 5 feet tall, and her risk for skin cancer is roughly doubled.

Q: What is next for research on this topic?

A: My work on human body size and cancer was driven by the need to confirm an important prediction of the multistage model. Another species in which this same effect is expected and appears to be strongly supported is the domestic dog: Larger dog breeds get more cancer than smaller-sized dogs. However, the more important future of this research is in the search to understand why the effect of size in increasing cancer is not seen when we compare a huge animal, such as an elephant, to a tiny animal, such as a mouse. Now that we can be confident that larger-sized individuals are more prone to cancer than smaller-sized individuals within a species, it is time to focus on comparative studies that involve species of different body size. What features of the biology of elephants and whales prevent body-size prediction from being true? We need to understand what adaptive changes have occurred in these large long-lived animals that suppresses their risk for cancer. If we can establish this, we may be able to use this information to develop therapeutic methods for minimizing cancer risk in humans.

PAGE BREAK

Q: Is there anything else that you would like to mention?

A: Traditional cancer biologists have done incredibly wonderful things in advancing our understanding of cancer and of therapies to control it, but I think there remains an important place for the comparative evolutionary approach. This approach has never been exploited in the past, but natural selection has clearly found ways of suppressing cancer in the largest-sized mammals. With an evolutionary approach, we are asking what is different about these animals, and if can they show us new avenues for controlling cancer. – by Jennifer Southall

References:

Green J, et al. Int J Epidemiol. 2011;doi:10.1093/ije/dyy065.

Nunney L. Proc Biol Sci. 2018;doi:10.1098/rspb.2018.1743.

Sung J, et al. Am J Epidemiol. 2009;doi:10.1093/aje/kwp088.

For more information:

Leonard Nunney, PhD, can be reached at University of California, Riverside, 900 University Ave., Riverside, CA 92521; email: nunney@ucr.edu.

Disclosure: Nunney reports no relevant financial disclosures.