4

A 2015 study finds a significant association between muscle-building supplements (MBS) and testicular germ cell carcinoma (TGCC) with seemingly massive odds ratios. There was a wave of panic about this study along with a lot of low-quality discussion (immediate dismissal due to grouping creatine and steroids, vows to immediately quit creatine, etc) on reddit and (unsurprisingly) supplement sites, but little after that. I think this study has a lot of public importance given rising testicular cancer and the prevalence of supplements, and that it's a huge shame that the vast majority of information you can find discussing it is so low-quality and biased.

The main thing that disturbs most people, including me, is the association with creatine, as risks with steroids are much more widely accepted (i.e., "is this the next asbestos?") There are a few things in particular I'm confused about.

To begin with, to summarize what I perceive to be the "relevant, important bits".

In the "Study population" section:

The eligibility criteria for cases in the study included having a histologically confirmed TGCC (Stage 0–IV) diagnosed during 2006–2010, no previous cancer diagnoses except for non-melanoma skin cancer, being a male resident of CT or MA and between the ages of 18–55 at diagnosis (...) Population-based controls were identified among English-speaking male residents of CT and MA between the ages of 18–55 at the time of the interview.

MBS use was defined as use for at least once a week for X4 consecutive weeks. The interview included an assessment of 30 different types of MBS powders or pills. The major ingredients, including creatine, protein, and androstenedione or its booster, were abstracted according to the product ingredients.

In the "Statistical analysis" section:

Unconditional logistic regression models were used to evaluate the associations between the use of MBS and the risk of TGCC. Odds ratios (OR) and 95% confidence intervals (CIs) were calculated for ever vs never MBS use and for several additional metrics in relation to risk of TGCC. These metrics included age at first use, number of MBS products used, and duration of use.

In the "Results" section:

We further conducted exploratory stratified analyses examining associations with TGCC for the major types of MBS use reported by the study population and found that the use of MBS containing ingredients of both creatine and proteins increased the risk of TGCC significantly (OR = 2.55, 95% CI: 1.05–6.15).

I think most discussion and confusion revolves around what exactly they asked the participants, and the implications and meaning of the last quote. I feel shaky on what they asked, and what "MBS containing ingredients of both creatine and proteins" means (in particular, that quote makes me wonder if they looked at creatine and protein individually and there was a very low OR so they selected the combo to make a point? Otherwise, how can this be interpreted and why did they word it in such a seemingly strange way?) Why would they choose to not disclose the list of substances they asked about?

Also:

In our study, nearly 20% of cases with TGCC had used MBS, which was similar to the previous case series study (Chang et al, 2005)

This is also confusing to me in the context of creatine being a major risk factor because since 2001 (TABLE 10-18c), young male creatine usage has fluctuated between 16~22%. Is that a valid counterargument? (I understand that's just looking at 12th graders, but if 22% of 12th graders, avg age 18, used it 5-10 years before the study and creatine prevalence hadn't shifted much for a few years before that, it doesn't seem too unreasonable to use that data to speculate non-cause for the study).

I've made a long post with a bunch of questions, and I could come up with many more, but ultimately, I want to form an educated opinion on whether it's reasonable to suspect that creatine causes an increased risk of testicular cancer, given the variety of benefits it has been studied to have. Any thoughts?

0

1 Answer 1

4
+50

Issues with the 2015 study

Demographics

First of all, any conclusions in the paper about the connection between creatine use and TGCC apply almost entirely to Caucasian males, not all males, as you can see from Table 1 that 95% of the cases in the study, and 89% of the control group, involved Caucasians:

Race Cases Controls
White 338 459
Other 18 54

Combining this with the fact that fewer than 19% of the case subjects and fewer than 13% of the control group ever used MBS (this amounts to only about 3 non-Caucasian TGCC patients in the study having ever used MBS, and only about 6 non-Caucasian control group members having ever used MBS, making no conclusions about non-Caucasians to be convincing here).

The study was not about creatine

The paper's claim is that there's a correlation between MBSs and TGCC, but creatine was just one of the 30 different types of MBSs included in the survey in which participants essentially said "yes" or "no" to questions about whether or not they had used MBSs, and for how long. The effect specifically from creatine was never isolated in the study. So the correlation between MBSs and TGCC was not necessarily caused by creatine, it could have (for example) have been caused by any one of the other 29 ingredients grouped together under the MBS umbrella for the purposes of the study.

Creatine was mentioned in the following quote (which turned out to also be part of the longer quote you included in your question):

"The interview included an assessment of 30 different types of MBS powders or pills. The major ingredients, including creatine, protein, and androstenedione or its booster, were abstracted according to the product ingredients."

After that, they never mentioned creatine again, except in the following quote (which it turns out, you also found):

"We further conducted exploratory stratified analyses examining associations with TGCC for the major types of MBS use reported by the study population and found that the use of MBS containing ingredients of both creatine and proteins increased the risk of TGCC significantly (OR = 2.55, 95% CI: 1.05–6.15)."

But the number "2.55" does not show up anywhere in the paper, as what is being mentioned here is supplementary to the main results reported in the paper. Usually the data behind a number like this, which is mentioned in the paper but without details about where it originated, would be described in the paper's "Supplemental Material", but this paper unfortunately did not have any. Again, this study was not about creatine, it was about MBSs, of which creatine was only 1 of 30, so it was not the goal of the authors to make any convincing conclusions about the impact of creatine (so those Reddit users were perhaps flawed in singling out creatine here).

Some other, more minor issues

The paper says:

"The prevalence of cryptorchidism and injury to the testes or groin was higher in cases than that in controls"

Each of these were about 10% higher, according to Table 1. Also, the paper didn't mention that the same table also shows in the row labeled "Family history of TGCC", that it was 2x higher in cases than controls. While the latter amounts to a minor issue on its own, combined with the 10% difference we see for cryptorchidism and for previous testes injury, there's a fair amount of other factors (specifically: prior cryptorchidism, testes injury, and family history of TGCC) that are significantly higher in the TGCC patients than in the control group.

Table 2 shows that the use of MBS was higher in the cases than in the controls, by less than 6%, which is smaller than the percentage of TGCC patients who had the other three pre-existing factors more than the control group. Combined with everything else said so far, this suggests that the 2015 paper may have been a good start to studying the connection between MBSs and TGCC, but a larger study would have to be done to answer your question.

Further reading

Since your question is about creatine, but the paper discussed throughout your entire question's body was about MBSs, for which creatine was only one of 30 and was never studied in isolation or at least with its own emphasis, I'll first recommend some work specifically about creatine, then some about TGCC, and then some about MBSs in general (some of these were done by the same group as the 2015 paper you described).

Specifically about creatine

Specifically about TGCC

About the effects of MBSs in general

1
  • 2
    Phenomenal quality answer, thank you so much! Jan 17, 2022 at 19:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.