ONLINE DRIVERS MEDIA

Dont fall for someone who overgeneralizes, misreports or -interprets study results to tell you that healthy mid-range testosterone levels were a major trigger of cancer development.

I just got an email from my good friend Carl Lanore who hosts the Super Human Radio Show I am sure you have already seen in the sidebar of the SuppVersity. He pointed me towards the results of a recent study from the Copenhagen University Hospital saying "Adel have you seen this? Im sure the media will be all over it." Since Carl is usually pretty good at identifying things that "news" make headlines, I would like to leapfrog the mass media craze before any of you consider cutting off their best parts to end up in the allegedly healthy depth of quasi zero testosterone.

You have no clue what I am talking about? Well, I am about to elaborate. You just have to stay with me for the rest of a comparatively long, but insightful (I promise) analysis of the study and related contemporary evidence.
So, where do I begin?  The "mass media compatible message" of the abstract (and this is usually not even as far as most headline producers read) says: Having normal testosterone levels increases your risk of dying from cancer before your time by at least 30%! The more testosterone, the likelier you will pass away before your expiry date."

Well, as I said, this is the mainstream interpretation. And interpretation anyone could identify as being fundamentally flawed by simply reading the abstract carefully. In fact, you dont even have to read the whole 268 words! It would suffice to read the end of the first line of the result section, where it says:

"For risk of early death after cancer, for men [...]" (Ørsted. 2014)

Did you notice it? There is an "after cancer" in this sentence. This means, your risk of dying, if you develop cancer is increased, if [whatever follows]. Now, that which follows is ...

• if you are in the 2nd quartile, your risk will be increased by 30%,
• if you are in the 3rd quartile, your risk will be increased by 31%,
• if you are in the 4th quartile, your risk will be increased by 52% and
• if you are in the 5th quartile, your risk will be increased by 80%.

So, if you happen to be unlucky enough to develop cancer and your testosterone levels are high there is in fact an increased risk you could die from cancer.

So testosterone is still bad, right? Yes, but anything that is "anabolic", i.e. promotes the growth of all cells in your body is "bad" for someone who has cancer. Guess what chemotherapy will to do your testosterone levels and the amount of other pro-anabolic factors in your body? It will wreak havoc on your testes (Wallace. 1997), reduce their size to that of dried raisins, increase your risk of gynecomastia and reduce your testosterone levels to exactly those 6-10 nmol/L (Whitehead. 1982) of which the previously cited study by Ørsted et al. found that they are associated with the least risk of dying from already existing cancer in your body.

Figure 1: If you look at a random assemble of the myriad of risk associations that have been established for low T, the results of the study at hand do no longer appear to be that frightening - right?

A result that conceals the established negative effects of having low testosterone (see Figure 1) or depressing it which androgen deprivation therapy which may reduce the testosterone levels to the allegedly desirable range, but is associated with a 350% (!) increased risk of dieing from cardiovascular disease in prostate cancer patients (Tsai. 2007). I guess this should make you reconsider the "usefulness" of low testosterone levels.

But there is also an increase in cancer risk, no?

Yes, there is. According to the scientists it is (I quote) "1.07 (95% CI 0.98–1.18) and 1.06 (0.93–1.22) for men and women" if you compare say a man with 10nmol/L to a man with 20nmol/L. Now, as far as I can remember a 95% confidence interval, which is what you see in brackets, i.e. for men 0.98-1.18, defines the range in which the chance that the hypothesis that is tested, i.e. "testosterone influences the risk of prostate cancer" has a 5% chance of not being bullocks... ah, I mean statistical significant.
So, the scientists are 95% sure that a 2x higher testosterone level will be associated with a 2% decrease and 18% increase in... does this ring a bell? Yeah, thats not exactly a reliable prediction considering the fact that the researchers adjusted for smoking status, cumulative smoking, body mass index, alcohol consumption, level of education, and level of income for men and women, but "forgot" to adjust for...

• family history of cancer, which is one of the, if not the main correlate of your risk of developing various cancer, such as prostate cancer (1000% increase, no typo | Steinberg. 1990), colon cancer (up to 59% risk increase depending on the region | Slattery. 1994) or breast cancer (145% risk increase with first-degree relative having breast cancer | Slattery. 1993) 
• diabetes, which has been found to be associated with a 60% increase in colorectal cancer risk in patients who have been diagnosed with diabetes 10+ years ago (La Vecchia. 1997), a
• the level of visceral fat, where high levels (relative to total body fat) are associated with 850% increased risk of breast cancer (Schapira. 1994) and up to 1000% increased prostate cancer risk (Hafe. 2004)
• low sleep duration and quality, which has been associated with an increased risk of developing almost every form of cancer you can think of (Blask. 2009), including 60% increased breast cancer risk for women working the "graveyard shift" (Davis. 2001)

I could go on with all sorts of nutritional factors, medication (specifically hormonal contraceptives) the amount of exercise, the area you live in, your year of birth and hundreds of other factors that have previously been associated with an increase of cancer risk, but I guess the above should suffice to make you less confident that a "confidence interval" of 0.98–1.18 in men and 0.93–1.22 in women was enough to make the claim that "having a high testosterone level would [mechanistically!] trigger the development of cancer.
References:

• Araujo, Andre B., et al. "Endogenous testosterone and mortality in men: a systematic review and meta-analysis." The Journal of Clinical Endocrinology & Metabolism 96.10 (2011): 3007-3019.
• Blask, David E. "Melatonin, sleep disturbance and cancer risk." Sleep medicine reviews 13.4 (2009): 257-264.
• Davis, Scott, Dana K. Mirick, and Richard G. Stevens. "Night shift work, light at night, and risk of breast cancer." Journal of the national cancer institute 93.20 (2001): 1557-1562.
• Garavello, Werner, et al. "Family history and the risk of oral and pharyngeal cancer." International journal of cancer 122.8 (2008): 1827-1831.
• Hafe, Pedro, et al. "Visceral fat accumulation as a risk factor for prostate cancer." Obesity research 12.12 (2004): 1930-1935.
• La Vecchia, Carlo, et al. "Diabetes mellitus and colorectal cancer risk." Cancer Epidemiology Biomarkers & Prevention 6.12 (1997): 1007-1010.
• Morgentaler, Abraham. "Testosterone and prostate cancer: an historical perspective on a modern myth." european urology 50.5 (2006): 935-939.
• Raynaud, Jean-Pierre. "Prostate cancer risk in testosterone-treated men." The Journal of steroid biochemistry and molecular biology 102.1 (2006): 261-266.
• Roddam, Andrew W., et al. "Endogenous sex hormones and prostate cancer: a collaborative analysis of 18 prospective studies." Journal of the National Cancer Institute 100.3 (2008): 170-183.
• Slattery, Martha L., and Richard A. Kerber. "A comprehensive evaluation of family history and breast cancer risk: the Utah Population Database." Jama 270.13 (1993): 1563-1568.
• Slattery, Martha L., and Richard A. Kerber. "Family history of cancer and colon cancer risk: the Utah Population Database." Journal of the National Cancer Institute 86.21 (1994): 1618-1626.
• Schapira, David V., et al. "Visceral obesity and breast cancer risk." Cancer 74.2 (1994): 632-639.
• Shores, Molly M., et al. "Low serum testosterone and mortality in male veterans." Archives of internal medicine 166.15 (2006): 1660-1665.
• Stattin, Pär, et al. "High levels of circulating testosterone are not associated with increased prostate cancer risk: a pooled prospective study." International journal of cancer 108.3 (2004): 418-424.
• Steinberg, G. D., Carter, B. S., Beaty, T. H., Childs, B. and Walsh, P. C. (1990), Family history and the risk of prostate cancer. Prostate, 17: 337–347. doi: 10.1002/pros.2990170409 
• Tsai, Henry K., et al. "Androgen deprivation therapy for localized prostate cancer and the risk of cardiovascular mortality." Journal of the National Cancer Institute 99.20 (2007): 1516-1524. 
• Ujpál, Márta, et al. "Diabetes and Oral Tumors in Hungary Epidemiological correlations." Diabetes care 27.3 (2004): 770-774.
• Wallace, Euan M., et al. "Effects of chemotherapy-induced testicular damage on inhibin, gonadotropin, and testosterone secretion: a prospective longitudinal study." The Journal of Clinical Endocrinology & Metabolism 82.9 (1997): 3111-3115.
• Whitehead, E., et al. "The effects of Hodgkins disease and combination chemotherapy on gonadal function in the adult male." Cancer 49.3 (1982): 418-422.

The advantage of HDL is its stability that reduces the risk of plaque build-up in the intestinal wall, which is clogged by the remnants of oxidized LDL and causes heart disease & co.

First things first: We are not talking about "hard experimental evidence" as you could generate it in a randomized controlled trial in a metabolic ward. The data I am reporting today is from a cohort with 1,566 participants with extensive lipid phenotype data completed the Harvard Standardized Food Frequency Questionnaire to determine their daily micronutrient intake over the past year - an epidemiological study that used stepwise linear regression was used to separately evaluate the effects of dietary covariates on adjusted levels of HDL-C, HDL-2, HDL-3, and apoA1.

Interestingly, this is the first study with a quality data-set that determined the association between specific dietary micronutrients with HDL-C, HDL-2, HDL-3, and apoA1, and how these dietary associations differ across the various measures of HDL - not just one.
To identify the HDL-promoters in the diet, the scientists use demographic and clinical variables in the base model. What they found was that numerous dietary intakes increased total HDL-C variance.
The results of their stepwise linear regression model in Table 1 indicate - probably for some people much surprisingly - that all alcohol intake levels were positively associated with HDL-C.

"In addition, magnesium, folate, and the saturated fat, myristic acid (14:0), were all positively and independently associated with HDL-C. Carbohydrate intake, iron, and % of fat derived from animal sources were each negatively additive for HDL-C." (Kim. 2014)

Similar effects from dietary intakes were observed for HDL-2, of which we know that it is decreased in women with rheumatoid arthritis (Arts. 2012) and individuals with other inflammatory diseases (including metabolic syndrom) and associated with a slightly higher reduction in acute myocardial infarction risk than "regular" HDL in several studies (Salonen. Salonen. 1991; Stampfer. 1991; Buring. 1992; Gaziano. 1993) for all alcohol intakes, magnesium, folate, and myristic acid (14:0), eicosapentaenoic acid (20:5, a ?-3 EPA), all of which were positively and independently associated with HDL-2 levels.

Table 1:  Best-fit model from stepwise linear regression predicting HDL-C levels using dietary intake data (Kim. 2014)

The opposite was the case for arachidonic acid (20:4, an ?-6 ARA), carbohydrate and iron intakes, which were both negatively associated with HDL-2 (see Table 1).

Dont forget to put things into perspective!

And just to make sure, I dont get angry emails from bulletproof coffee drinkers: Your coffee is fine, the content of the only "good" saturated fat, i.e. myristic acid, happens to be especially high coconut oil (41%; Sodamade. 2013) and relatively high in butter (12% independent of whether its grass-fed or not; Couvreur. 2006) - surprised? Not really, I guess. As a SuppVersity reader you are by now aware that the bad reputation coconut oil and butter have for being mostly saturated fats is no longer supported by contemporary scientific evidence (Dias. 2014).

And with respect to the total animal fat intake - for the average Westerner thats a good measure of how much processed meat he / she eats, so I would not overrate the small negative association (1/80 of the one of having a ton of carbohydrates in your diet!) the scientists found in the study at hand.
References:

• Arts, Elke, et al. "High-density lipoprotein cholesterol subfractions HDL2 and HDL3 are reduced in women with rheumatoid arthritis and may augment the cardiovascular risk of women with RA: a cross-sectional study." Arthritis Res Ther 14.3 (2012): R116.
• Buring, J. E., et al. "Decreased HDL2 and HDL3 cholesterol, Apo AI and Apo A-II, and increased risk of myocardial infarction." Circulation 85.1 (1992): 22-29.
• Couvreur, S., et al. "The linear relationship between the proportion of fresh grass in the cow diet, milk fatty acid composition, and butter properties." Journal of dairy science 89.6 (2006): 1956-1969.
• Dias, C. B., et al. "Saturated fat consumption may not be the main cause of increased blood lipid levels." Medical hypotheses 82.2 (2014): 187-195.
• Gaziano, J. Michael, et al. "Moderate alcohol intake, increased levels of high-density lipoprotein and its subfractions, and decreased risk of myocardial infarction." New England Journal of Medicine 329.25 (1993): 1829-1834. 
• Kim et al. "Effects of dietary components on high-density lipoprotein measures in a cohort of 1,566 participants." Nutrition & Metabolism 2014, 11:44.
• Leon, ARTHUR S., and OTTO A. Sanchez. "Response of blood lipids to exercise training alone or combined with dietary intervention." Medicine and science in sports and exercise 33.6 Suppl (2001): S502-15.
• Salonen, Jukka T., et al. "HDL, HDL2, and HDL3 subfractions, and the risk of acute myocardial infarction. A prospective population study in eastern Finnish men." Circulation 84.1 (1991): 129-139. 
• Sodamade, A¹, and O. S. Bolaji. "Fatty acids composition of three different vegetable oils (soybean oil, groundnut oil and coconut oil) by high-performance liquid chromatography." Chemistry and Materials Research 3.7 (2013): 26-29.
• Stampfer, Meir J., et al. "A prospective study of cholesterol, apolipoproteins, and the risk of myocardial infarction." New England Journal of Medicine 325.6 (1991): 373-381.

The study at hand used plain ascorbic acid, no quack supplements with "advanced vitamin C".

While people tend to believe that vitamin C is good for anything, the evidence that it actually does anything good is relatively scarce. Against that background I am happy to tell you that a group of Greek researchers from the School of Physical Education and Sport Science, the European University Cyprus and theAristotle University of Thessaloniki have now finally confirmed what many of you probably thought was a long-established fact: "[L]ow vitamin C concentration is linked with decreased physical performance and increased oxidative stress and that vitamin C supplementation decreases oxidative stress and might increase exercise performance only in those with low initial concentration of vitamin C." (Paschalis. 2014)
When they came up with the study design, Paschalis et al. simply assumed that the mythical ergogenic effect of vitamin C actually existed. To test this hypothesis, they screened 100 males for vitamin C baseline values in blood, picked the 10 individuals with the lowest and the 10 with the highest vitamin C values from their baseline sample and assigned them to two groups.

Figure 1: Overview of the study design (Paschalis. 2014)

Using a placebo-controlled crossover design, the 20 selected subjects performed aerobic exercise to exhaustion (oxidant stimulus) before and after vitamin C supplementation for 30 days.

An overview of the study design is shown in Fig. 1. All measurements were performed between 08:00 and 11:00 h after overnight fasting. Initially, to examine whether rest ing blood vitamin C concentration affects aerobic perfor mance, VO2max was assessed (using incremental cycling test to volitional exhaustion) and was compared in both the low and the high vitamin C groups (Monark, Vansbro, Swe den). More specifially, the protocol started with a 50 W load at 50 rpm and increased by 10 W every 2 min until volitional fatigue. The test was terminated when three of the following four criteria VO2max were met: (1) volitional fatigue, (2) a lower than 2 mL/kg/min increase in VO2 despite an increase in workload, (3) a respiratory exchange ratio greater than or equal to 1.10, and (4) heart rate within 10 bpm of the predicted maximal heart rate (220–age). Res piratory gas variables were measured using a metabolic cart (Quark b2, Cosmed, Italy), which was calibrated before each test using standard gases of known concentration. The VO 2max assessment was used as a reference value to cal culate the workload at the relative intensity of each subject and ensured that all subjects would cycle at similar relative intensity during the following aerobic exercise sessions.

After the baseline testing had been done, the subjects within both the low and the high vitamin C groups received either placebo (3x333mg of lactose) or vitamin C supplementation (3x333mg of vitamin C), in a double-blind randomized crossover fashion (see Figure 1).

Figure 2: Changes in VO2max (left) and redox status (right) in subjects according to initial vitamin C status before and after vitamin C supplementation for 30 days (Paschalis. 2014).

As you can see in Figure 2 there were measurable differences in the response to the acute exhaustive exercise protocol (an oxidant stimulus), the subjects in both groups performed before and after vitamin C or placebo supplementation for 30 days. The data in Figure 2 does yet also show that the subjects who had been randomly assigned to the vitamin C supplement group had lower baseline VO2max levels. A fact that raises the question whether this is the result of a lower vitamin C intake or whether the vitamin C intake correlates with an unhealthier lifestyle that left the subjects unfit and with low vitamin C levels.
Unfortunately, this question is hard to answer based on the available research on vitamin C. While we have conflicting results with respect to its ability to impair the adaptational response to exercise (Close. 2014), there is very little evidence that it will actually have beneficial effects on any meaningful performance parameters. In fact, a study by Huck et al. that was published in the scientific journal Nutrition in 2013 is probably what comes closest to the results of the study at hand.

Figure 3: Effects of 500mg vitamin C per day on selected parameters in a 4 week chronic exercise + diet supplementation in obese men and women (Huck. 2013)

In said study Huck et al. observed that the provision of 500mg of vitamin C as an adjunct to exercise and diet in obese individuals lead to significant reductions in heart rate and the ratings of perceived exertion during exercise. The data in in Figure 3 does yet also tell you that there were no beneficial effects on VO2max, which best reflects the adaptational response to exercise.

This results of stands in contrast to the study at hand, but in line with previous results of studies in athletes, where only more or less irrelevant reductions of the acute inflammatory response to exercise were observed (Nieman. 2000; Peters. 2001; Tauler. 2002). A response of which you as a SuppVersity reader know that it is an essential part of the signalling cascade that triggers the adaptational response to. If we eventually get back to the Paschalis study, it would thus appear that athletes who are usually consuming more than enough vitamin C in their diets and are not at particular risk of developing low serum vitamin C levels would see similar results as the "high vitamin C" subjects in the Paschalis study, i.e. none - even worse, in view of the potential negative effects on the training induced adaptations that could not become visible in the study at hand, because there was no exercise protocol involved, it could even harm their progress.
References:

• Close, G. L., and M. J. Jackson. "Antioxidants and exercise: a tale of the complexities of relating signalling processes to physiological function?." The Journal of physiology 592.8 (2014): 1721-1722.
• Huck, Corey J., et al. "Vitamin C status and perception of effort during exercise in obese adults adhering to a calorie-reduced diet." Nutrition 29.1 (2013): 42-45.
• Nieman, David C., et al. "Influence of vitamin C supplementation on cytokine changes following an ultramarathon." Journal of Interferon & Cytokine Research 20.11 (2000): 1029-1035.
• Paschilis, V. et al. "Low vitamin C values are linked with decreased physical performance and increased oxidative stress: reversal by vitamin C supplementation." Eur J Nutr (2014). Ahead of print.
• Paulsen, Gøran, et al. "Vitamin C and E supplementation hampers cellular adaptation to endurance training in humans: a double?blind, randomised, controlled trial." The Journal of physiology 592.8 (2014): 1887-1901.
• Peters, E. M., et al. "Vitamin C supplementation attenuates the increases in circulating cortisol, adrenaline and anti-inflammatory polypeptides following ultramarathon running." International journal of sports medicine 22.7 (2001): 537-543.
• Picklo, Matthew. "Supplementation with vitamin E and vitamin C inversely alters mitochondrial copy number and mitochondrial protein in obese, exercising rats (1030.5)." The FASEB Journal 28.1 Supplement (2014): 1030-5. 
• Powers, Scott K., et al. "Dietary antioxidants and exercise." Journal of sports sciences 22.1 (2004): 81-94.
• Powers, Scott K., And Kurt J. Sollanek. "Endurance Exercise And Antioxidant Supplementation: Sense Or Nonsense?-Part." Sports Science 27.137 (2014): 1-4.
• Tauler, P., et al. "Diet supplementation with vitamin E, vitamin C and ?-carotene cocktail enhances basal neutrophil antioxidant enzymes in athletes." Pflügers Archiv 443.5-6 (2002): 791-797.