Editor’s Spotlight/Take 5: Are There Sex Differences in the Association of Alcohol Consumption With the Risk of Soft Tissue Sarcoma? A Nationwide Population-based Study in Korea

Min Wook Joo MD, PhD

Chocolate is bad for you. No, chocolate is good for you. Working while seated at a desk gives you “sitting disease” [14], which may kill you [13]. No, standing at your desk for 3 hours a day burns no more than the caloric equivalent of a carrot [15].

Are they coming for my whiskey this month in Clinical Orthopaedics and Related Research®? Let’s find out.

The Editor’s Spotlight/Take 5 article this month, “Are There Sex Differences in the Association of Alcohol Consumption With the Risk of Soft Tissue Sarcoma? A Nationwide Population-based Study in Korea” [8], is the product of a multicenter collaboration from both sides of the Pacific Ocean. Its authors plumbed a robust national database from South Korea to discover that although alcohol consumption did not increase the risk of developing soft tissue sarcoma in males, it did among females, with dose-response relationships appearing in terms of the amount of drinking done on a per-occasion basis, especially at the higher end of alcohol consumption in women.

This seems important for two reasons. First, the finding itself, if confirmed, appears to justify public health–oriented efforts to raise awareness of this newly discovered risk of drinking, at least in women, because the hazard ratios were large: Among women who engage in binge drinking (> 14 drinks at a session), the hazard was estimated at 370% [8].

But that isn’t the only reason I thought this paper was important enough to feature in this section of CORR®. And I almost didn’t feature it—after all, a study of this design can’t prove cause-effect, only association, and the other issue is that because soft tissue sarcomas are rare, even large risk increases may not correspond to a large number of excess sarcomas (though one could say even one excess sarcoma is one too many).

But there was a good reason to put the Spotlight on this paper [8], and there’s a good reason for you to read it: Had the authors not subanalyzed their dataset by sex, a potentially important finding would have been altogether missed. In the overall group containing both males and females, there was no relationship; it was only when the authors looked at males and females separately that they discovered that women who consumed more alcohol were at greater risk of developing a soft tissue sarcoma.

The issue of paying careful attention to how sex and gender are treated as variables in scientific reporting is one we’ve emphasized heavily in CORR over a long stretch of time. The editors here first articulated our standards on it more than 10 years ago [11], some of which were rolled into the International Committee of Medical Journal Editors’ reporting recommendations document [6], as well as in the more recent Sex and Gender Equity in Research (SAGER) guidelines [5]. We solicit the best available research on sex, gender, and other disparities in an every-other-year symposium we publish on the topic; the one last year was great [16], and we’re collecting work for the next one now, so please do email me if you have something to contribute. And most recently, we worked with a large group of orthopaedic surgery journal editors who are committed to improving how the leading journals of our specialty handle this important topic [12]. Doing so, we think, will help both men and women be healthier.

But for now, let’s talk about a somewhat surprising harm of excess alcohol consumption among women, as well as why reporting thoughtfully by sex and gender is so important. Join me as I go behind several fascinating discoveries with Min Wook Joo MD, PhD, the senior author of “Are There Sex Differences in the Association of Alcohol Consumption With the Risk of Soft Tissue Sarcoma? A Nationwide Population-based Study in Korea” [8], from St. Vincent’s Hospital at The Catholic University of Korea in Seoul.

Take 5 Interview with Min Wook Joo MD, PhD, senior author of “Are There Sex Differences in the Association of Alcohol Consumption With the Risk of Soft Tissue Sarcoma? A Nationwide Population-based Study in Korea”

Seth S. Leopold MD: Congratulations on this thought-provoking study. First, how “big” is this finding—why do you think it’s important from the public health standpoint, given the relative rarity of soft tissue sarcomas? Do your findings hint at other public health directions we should be thinking about?

Min Wook Joo MD, PhD: While some might argue soft tissue sarcomas shouldn't be a top priority in public health because they are rare, alcohol is already known to contribute to many common diseases. Our finding of alcohol being associated with the development of soft tissue sarcomas strengthens the public health rationale for moderation by providing another layer of evidence supporting efforts to reduce alcohol consumption. This finding should inspire a shift in thinking: While many common diseases have well-established causes and are already widely recognized by the public, rare diseases linked to modifiable risk factors warrant increased public health attention because some of those risk factors likely contribute to hundreds of other conditions. Additionally, when you consider how rare cancers typically attract limited attention and investment from the private sector, despite their devastating long-term impacts, a strong case should be made for national-level intervention and support through public health strategies. It is essential that we continue advocating for policy change through persistent and evidence-based efforts.

Dr. Leopold: How confident should we be in a data source when the social variable in question—here, alcohol use—has two strikes against it: It was self-reported, and there’s often an element of shame associated with its use (or with reporting its use)?

Dr. Joo: You've touched on a very important issue. Since alcohol consumption was collected through self-reporting, there's certainly a possibility of underreporting. That said, underreporting tends to bias the effect estimates toward the null, meaning the true association might actually be stronger than what’s observed. So if we still see a meaningful result despite that, I think this limitation actually strengthens the credibility of our finding.

Dr. Leopold: Given the two concerns above, what would be the best setting in which to replicate, extend upon (perhaps with other kinds of cancer), or refute your main findings, ideally in a way that might confirm the idea that this is a cause-effect relationship? What should that line of research look like and how should it best be performed?

Dr. Joo: To improve the reliability of exposure measurement, efforts to assess alcohol consumption more objectively would be valuable. For example, using biomarkers or adding objective indicators to complement self-reported data could help validate the reported intake. However, since implementing this is not always feasible in practice, I believe that large-scale, longitudinal studies such as ours [8] remain the most practical and appropriate approach to estimate a potential causal relationship. This design helps reduce bias from underreporting while preserving the ethical and practical strengths of a cohort study.

Dr. Leopold: Shifting gears a bit: Had you not disaggregated and analyzed your dataset by sex, you’d have missed some very important findings. Why did you feel it was so important to look separately at males and females? Did similar studies about other cancers find big differences between the sexes that caused you to wonder?

Dr. Joo: Certainly, the observation of sex differences in many other cancer studies raised similar questions in our research. I was particularly inspired by the article, “Sex and Tumor-Site Differences in the Association of Alcohol Intake With the Risk of Early-onset Colorectal Cancer” [7], which showed how global per-capita alcohol consumption has been increasing and may be contributing to a growing burden of cancer worldwide. We also identified a more specific concern in the Korean context: From 2005 to 2018, the prevalence of high-risk drinking—defined as drinking at least twice a week and consuming a large amount of alcohol per occasion (which is an average of 7 or more glasses for men and 5 or more glasses for women)—sharply escalated among women, from 3.4% in 2005 to 8.4% in 2018 [9]. This rate dipped to 6.9% in 2021, but data from this most recent year continued to show a higher prevalence of high-risk drinking among women in their 20s (9.5%) and 30s (10.3%) compared to other age groups [9]. This marked upward trend in women prompted us to pay closer attention to potential sex differences in our analysis. Finally, the Department of Statistics and Actuarial Science at Soongsil University in Seoul, South Korea has conducted statistical analyses for over 1000 studies using the National Health Insurance Service (NHIS) database, and those analyses have consistently identified sex-specific differences in exposure-response relationships across numerous studies. I would like to take this opportunity to acknowledge Kyungdo Han PhD and Ye Bin Park MSc in this department for their invaluable contributions to the statistical analysis and interpretation in this study [8].

Dr. Leopold: To what do you attribute the difference you found between the sexes?

Dr. Joo: Females face a greater risk of alcohol-related cirrhosis and other liver diseases. Excessive drinking heightens the risk of heart muscle damage in females at lower consumption levels and over a shorter duration. Additionally, alcohol-related cognitive decline and brain shrinkage occur earlier. Females eliminate a similar amount of alcohol per unit body weight as males [10] but more alcohol per unit of lean body mass per hour [1, 4]. However, females’ lower percentage of body water leads to higher blood alcohol concentrations compared to males with similar body weight, even after consuming the same quantity of alcohol [3, 4]. Moreover, the smaller distribution volume of alcohol in females is linked to prolonged periods of elevated ethanol blood levels [2]. Thus, females may have higher vulnerability to the effects of alcohol compared to males [4].