Strength & Cancer-related Mortality: Is it enough to exercise, or do you actually need to get strong?

Cancer is among the leading causes of disease and death in the U.S. and contributes to a substantial burden of disease in the world [CDC, WHO]. Enormous amounts of resources are spent on treatment as well as research into new treatment approaches (e.g., chemotherapeutics, immunotherapies, and other emerging treatments) that often provide small, incremental survival benefits over existing therapies at great cost. [Prasad 2015]

Key Points:

1. Among patients with cancer, sarcopenia is prevalent at all stages of disease. It is a strong independent predictor of both cancer-related and all-cause mortality, independently increases the risk of treatment complications, and is associated with cancer-related fatigue, pain, and quality of life.
2. In this study of a nationally-representative sample of 2,773 adults, being in the top quartile for knee extensor strength was associated with an approximately 50% risk reduction for cancer-specific mortality after controlling for co-variates. These data suggested that approximately 20% of deaths due to cancer were attributable to not being in the top quartile for knee extension strength.
3. In contrast, simply engaging in muscle-strengthening activities was not associated with significant reduction in cancer-specific mortality. In other words, it appears that it is not enough to simply exercise, but rather that in order to enjoy maximal risk reduction, one must actually get strong.

Among physicians and oncologists, the end-stages of cancer-related wasting syndromes (known as cachexia) are well-recognized as poor prognostic factors associated with increased risk of treatment complications and mortality. [Tisdale 2002] However, the prevalence and significance of sarcopenia in earlier-stage disease is under-recognized. [Christensen 2014] It is also less appreciated whether inexpensive, accessible lifestyle interventions to counteract this progressive loss of muscle mass can attenuate cancer-related risks. To address this important question, I’ll review a 2018 paper by Dankel et al, “Cancer-Specific Mortality Relative to Engagement in Muscle-Strengthening Activities and Lower Extremity Strength.”

Purpose, Subjects, and Methods

The authors sought to use a nationally representative sample of U.S. adults to analyze the effects of 1) skeletal muscle strength and 2) engagement in muscle-strengthening activities on the overall risk of death due to cancer.

Data were obtained from the 1999-2002 National Health and Nutrition Examination Survey (NHANES), which specifically included measures of knee extensor strength.

This included a total of 2,773 individuals aged ≥ 50 years, 50.4% female, and 58% non-Hispanic whites.

The analysis used 1) measures of lower extremity strength, 2) rates of engagement in muscle-strengthening activities, and 3) cancer-specific mortality, as well as numerous co-variates that were used for adjustment. These data were obtained as follows:

1. Lower extremity strength was measured using an isokinetic dynamometer. Participants performed 3 warm-up repetitions followed by 3 maximal isokinetic contractions at a speed of 60 degrees per second. The peak force produced over the three repetitions was measured and corrected for gravity. [Validity]
2. Engagement in muscle-strengthening activities was measured via individual self-report in response to the following questions: “During the past 30 days, did you do any physical activities specifically designed to strengthen your muscles, such as weight lifting, push-ups, or sit-ups?” and, if so, “During the past 30 days, how many times did you do these muscle strengthening activities (e.g., weight lifting, push-ups, or sit-ups)?”. Individuals reporting performance of at least 8 sessions within the prior month (i.e., an average of two sessions per week) were classified as engaging in muscle-strengthening activities. [Validity]
3. Cancer-specific mortality was determined by matching personal identification information with the National Death Index (NDI), followed by manual examination of medical records and associated International Classification of Disease (ICD) coding.

Co-variate data obtained and used for adjustment purposes included self-reported aerobic activity, age, race, total blood cholesterol, mean arterial blood pressure, body mass index, serum C-reactive protein, reported smoking status, and reported use of ambulatory devices (e.g., a cane), statin medications, arthritis, congestive heart failure, coronary artery disease, cancer, diabetes, and stroke.

The data were analyzed using knee extensor strength (dichotomized to the 75th percentile and above vs. below this cutoff) and engagement in muscle-strengthening activities (dichotomized to 8+ sessions per month vs. less than 8 sessions per month) as the independent variables of interest.

The authors then performed three analyses: a minimally-adjusted model (using only age, sex, race, and aerobic activity as co-variates), an extended adjusted model (using all of the above-listed co-variates), and an additional adjusted model comparing the bottom quartile (i.e., 25th percentile) versus the upper three quartiles, rather than the upper quartile (i.e., 75th percentile) versus the bottom three quartiles.

Findings:

Overall, 382 individuals (13.8%) of the sampled population met sufficient engagement in muscle-strengthening activities (i.e., 8+ sessions per month), while 1009 (36.4%) met aerobic physical activity guidelines.

Of the 699 individuals in the 75th percentile for knee extensor strength, 32 (4.5%) died of cancer, compared to 130 (6.6%) of the 1,944 individuals below this cutoff. Being in the top quartile for knee extensor strength was associated with a 53% risk reduction for cancer-specific mortality in the minimally-adjusted model, and a 50% risk reduction in the extended adjusted model. 

Analyzing the data by sex showed that for every 15 N increase in knee extension strength, men had a 5% and women had an 8% reduced risk of cancer-specific mortality (Men, HR 0.95, 95% CI 0.91-0.99, P=0.01; Women, HR 0.92, 95% CI. 0.86-0.91, P=0.01).

There was no evidence of interaction effects between strength and age, sex, baseline history of cancer, body mass, or aerobic activity. In fact, excluding the 394 individuals who had ever been diagnosed with cancer from the study cohort, strength maintained a strong inverse association with cancer-specific mortality (HR 0.43, 95% CI 0.22-0.84, p=0.01).

Interestingly, when “flipping” the analysis to examine the bottom 25th percentile versus the upper three quartiles, the results were no longer significant (discussed further below).

In comparison to these data on knee extension strength, engagement in muscle-strengthening activities was associated with a 6% risk reduction for cancer-specific mortality in the minimally-adjusted model, and an 8% risk reduction in the extended adjusted model. However, neither of these findings were statistically significant.

A final analysis computed a statistic known as the Population Attributable Fraction, which aims to describe the proportion of cancer-specific mortality that can be attributed to a specific variable; in this case, strength. The PAF for those in the 75th percentile versus below was estimated at 20.9%, suggesting that approximately 20.9% of deaths due to cancer are attributable to not being in the top quartile for strength. Theoretically, this means that about one out of every five cancer deaths could have been averted if the individuals had been in the top quartile for strength.

Take-Home Message

Cancer is known to have profoundly catabolic effects and leads to a generalized cancer-related wasting syndrome known as cachexia in its end stages. Sarcopenia (the loss of muscle mass and strength), however, can be present at all stages of disease and often goes unrecognized in earlier stages. For example, Burden et al found that 54% of newly diagnosed early-stage colorectal cancer patients had a handgrip strength below 85% of the reference range for healthy age-matched controls. [Burden 2010] Similarly, a study of 714 newly diagnosed patients with a variety of cancers found that they carried an average of 0.9 kg less muscle mass compared with healthy controls prior to the initiation of any treatment. [Cao 2010]

This decrease in muscle mass and strength occurs early and progresses over time through multiple complex mechanisms. These include things like tumor-derived systemic inflammation, chemotherapy and other drug-related effects, and lifestyle-related factors such as physical inactivity and malnutrition. These collectively  induce a state of anabolic resistance, whereby an individual demonstrates a blunted (or absent) response to a given dose of anabolic stimulus. Practically speaking, this means  they are less “sensitive” to a given dose of protein or exercise. Additionally, these complex mechanisms promote catabolism of lean body mass.

There has been increasing research examining the role of skeletal muscle strength and function in cancer-related outcomes. [Christensen 2014] For example, skeletal muscle mass and function are strong independent predictors of both cancer-related and all-cause mortality. [Ruiz 2010] Additionally, sarcopenia independently increases the risk of treatment complications such as dose-limiting toxicity from chemotherapy and surgical complications (including death). Finally, there are strong associations between sarcopenia and patient-reported outcomes such as cancer-related fatigue, pain, and quality of life.

This study was the first to use a nationally representative sample of U.S. adults to analyze the effects of skeletal muscle strength and engagement in muscle-strengthening activities on the overall risk of death due to cancer. Limitations include the retrospective design, self-report of engagement rates in strength and aerobic exercise, and the relatively low mortality rate in the studied population (160 of 2,773 adults). Additionally, while the authors performed adjusted analyses using a number of co-variates (as described above), there may be additional unmeasured variables that were unaccounted for in the present analysis.

The authors found that being in the top quartile for knee extensor strength was associated with a 50% reduced risk for cancer-specific mortality after adjustment for a number of co-variates. However, there was no significant risk reduction associated with engagement in muscle-strengthening activities alone. This means that simply participating in these sorts of activities alone is not sufficient to earn the mortality benefit – one must actually get strong in order to enjoy these benefits.

Furthermore, when authors “flipped” their analysis to examine the bottom quartile (i.e., 25th percentile) versus the upper three quartiles, the results were no longer significant. This suggests that simply avoiding being in the bottom quartile is not enough to achieve maximal risk reduction; again, one must actually get strong.

Notably, this presents a challenge given the wide inter-individual variability in baseline strength and in response to strength training interventions (see April 2019 BMR). [Ahtiainen et al. 2016] For an individual with low baseline strength and with a poor response to a particular strength training intervention, they could plausibly be at an increased risk of cancer-specific and/or all-cause mortality.

It is therefore important to recognize those with low physical strength and provide appropriately-dosed interventions to improve muscle mass and muscle function. We have evidence suggesting that resistance training can attenuate or reverse cancer-induced anabolic resistance. [Montalvo 2018] However, this anabolic resistance can be progressive, and end-stage cachexia represents a stage where patients may become nearly refractory to such anabolic stimuli. [Antoun 2018] Therefore, resistance training and nutrition interventions should occur as early as possible. Additionally, the “dose” of these interventions likely require titration on an individual basis over time in order to continue generating the desired adaptations.

While we have a good understanding of how resistance exercise and nutrition interventions can generate improvements in skeletal muscle mass and function, the specific mechanisms by which such interventions exert their beneficial effects on cancer outcomes are likely complex and multifactorial, and as of now remain poorly understood. Similarly, exactly how strong is strong enough for these health outcomes remains unknown as well, and may ultimately prove to be a highly individual threshold. While all of this will require additional research to clarify, we can still feel confident in recommending strength training interventions to patients with cancer to reduce their risk of mortality.

To summarize, clinicians and patients should understand: 

1. 1. the significance of sarcopenia with respect to cancer-related outcomes, 
   2. that simply engaging in activity (or “being active”) is not enough to obtain maximal risk reduction, and 
   3. that one should actually get stronger to maximize benefit.

References:

1. Dankel et al. Cancer-Specific Mortality Relative to Engagement in Muscle-Strengthening Activities and Lower Extremity Strength. J Phys Act Health. 2018 ;15:144-149.
2. Burden et al. Nutritional status of preoperative colorectal cancer patients. J Hum Nutr Diet 2010; 23: 402–407.
3. Cao et al. Resting energy expenditure and body composition in patients with newly detected cancer. Clin Nutr 2010; 29: 72–77. 
4. Montalvo et al. Resistance Exercise’s Ability to Reverse Cancer-Induced Anabolic Resistance. Exerc. Sport Sci. Rev., Vol. 46, No. 4, pp. 247–253, 2018 
5. Ruiz et al. Association between muscular strength and mortality in men: prospective cohort study. BMJ 2008; 337: a439. 
6. Antoun et al. Muscle protein anabolism in advanced cancer patients: response to protein and amino acids support, and to physical activity. Annals of Oncology 29 (Supplement 2): ii10–ii17, 2018 
7. Ahtiainen et al. Heterogeneity in resistance training-induced muscle strength and mass responses in men and women of different ages. Age (2016) 38:10
8. Tisdale. Cachexia in cancer patients. Nat Rev Cancer. 2002 Nov;2(11):862-71.
9. Christensen et al. Muscle dysfunction in cancer patients. Ann Oncol. 2014 May;25(5):947-58.