Abstract
Background/Aim: Skeletal muscle mass loss is an emerging concern in oncology. Our systematic review and meta-analysis identified the mean difference in skeletal muscle index pre- to post-chemotherapy and synthesized potential key factors. Materials and Methods: We searched primary original research published through October 2019 in four databases: MEDLINE via PubMed, Scopus, CINAHL, and Embase. Results: Fifteen studies were included, 60% published in the past 2 years (2018-2019). Advanced non-small cell lung cancer was the most frequently reported cancer, and overall survival the most often identified key related factor. Mean difference in skeletal muscle index during chemotherapy was 2.72 (95%CI=1.77-3.67, p=0.00), with muscle loss in males (4.52, 95%CI=3.34-5.71, p=0.00) about 1.6 times higher than that in females (2.86, 95%CI=0.81-4.92, p=0.01). Conclusion: Oncologists should recognize sex-specific differences in skeletal muscle mass loss during chemotherapy and consider adjusting treatment accordingly.
Low skeletal muscle mass is an emerging issue in oncology. While the amount of skeletal muscle mass loss varies widely across cancer types, between 5% and 89% of cancer patients have low skeletal muscle mass (1). Low skeletal muscle mass at baseline has been reported to the increase incidence of disability among cancer patients and is associated with poor anti-tumor response (2). In addition, cancer patients with low skeletal muscle mass during cancer treatment have been reported to have a higher risk of mortality (3-5), cancer recurrence (6-8), and reduced quality of life (9).
The causes of low skeletal muscle mass are multifactorial. These include cancer itself, cancer treatments, and aging (10). Chemotherapy is known to accelerate muscle mass loss in cancer patients. For example, gastric cancer patients who received adjuvant chemotherapy had significantly decreased skeletal muscle mass, which was an independent risk factor for overall survival rate (11). In addition, higher risk of chemotherapy toxicity has been related to low muscle mass, with 50% of metastatic breast cancer patients with sarcopenia, compared to 20% of those without sarcopenia, showing capecitabine toxicity (12). Since chemotherapy is linked to loss of muscle mass, which is in turn linked to an increased risk of mortality (12), cancer specialists need to recognize the importance of skeletal muscle mass as a predictor of survival rates.
Recent studies have offered reasons for viewing skeletal muscle mass change as an essential predictor of survival rates, regardless of whether a patient meets criteria for sarcopenia before or after treatment. In a previous retrospective study involving 394 patients with nasopharyngeal carcinomas, the presence of sarcopenia, whether before or after cancer treatment, was not related to overall survival. Severe muscle loss after chemotherapy, however, was an independent predictor of prognosis (13). In another retrospective study involving locally advanced cervical cancer patients, pretreatment sarcopenia was not related to overall survival. However, skeletal muscle mass loss during concurrent chemoradiation therapy was independently related to both poorer overall survival and cancer-specific survival (14). Past systematic review and meta-analysis papers have focused on low muscle mass either before treatment or after treatment. We found no such studies examining skeletal muscle mass change during chemotherapy. The focus of our meta-analysis thus reflects researchers' current areas of interest regarding skeletal muscle mass change and examines potential key factors related to skeletal muscle mass loss as well as mean differences in skeletal muscle mass change during treatment.
PRISMA flow diagram.
Materials and Methods
Data sources and search strategy. We performed a comprehensive search for relevant articles published from 1973 through October 2019 using four databases: MEDLINE via PubMed, Scopus, the Cumulative Index to Nursing and Allied Health Literature (CINAHL) Plus with Full Text, and the Excerpta Medica Database (Embase). Because the databases use differing Medical Subject Headings (MeSH) terms, a research librarian and expert researchers helped to identify MeSH terms likely to produce the most accurate search results. The general search terms were neoplasms (or cancer or oncology or cancer survivor) AND skeletal muscle index (or low muscle mass or sarcopenia or muscular atrophy) AND chemotherapy. Only English-language published articles for which full-text versions were available were included. Figure 1 illustrates the detailed search strategy employed. This systematic review and meta-analysis were guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Titles and abstracts were independently reviewed by two reviewers, and the full text of each relevant article was obtained. After selecting relevant studies, the two reviewers assessed their quality and extracted appropriate data.
Study selection. The following criteria were used to select studies for inclusion: (a) primary original research published in a journal; (b) a study sample consisting of cancer patients and survivors; (c) a study that included repeated measurements of skeletal muscle index using computed tomography (CT); and (d) a study that appeared in a published English-language article. We excluded studies that: (a) did not relate to cancer, sarcopenia, or muscle mass change; (b) were animal studies; (c) did not report original primary research (such as review papers); (d) did not report skeletal muscle index (such as those using different units or measurements); (e) applied interventions that produced muscle mass change (such as exercise and nutritional interventions); (f) did not include repeated muscle mass measurements during chemotherapy; (g) focused on skeletal muscle mass change related to surgery or radiation therapy; or (h) were not suitable for meta-analysis because median values and ranges were reported rather than mean and standard deviation (SD) or standard error (SE) values.
Data extraction and analysis. The data that were collected from each of the 15 included articles of the study sample included study characteristics, measurements of muscle mass (mean and SD, or mean and SE), mean time between CT assessments, and key findings associated with skeletal muscle mass. To assess the quality of the nonrandomized studies in this review, the Newcastle–Ottawa Scale was applied independently by two reviewers (15). Three categories, including selection of the study population, comparability, and description of the outcome, were assessed, and high-quality studies were defined as those meeting 7 or more out of 9 items.
To test for heterogeneity of the studies, we calculated I2 statistics and Cochrane's Q statistics. The I2 value is the ratio of the interstudy variance. If an I2 value exceeded 50% and the p-value of χ2 was below 0.1, we concluded that there was substantial heterogeneity according to the criteria suggested in the Cochrane Handbook for Systematic Reviews of interventions (16). To estimate mean differences for parallel group analysis, the skeletal muscle mass change values reported across the studies were used. Specifically, mean differences between skeletal muscle index (cm2/m2) values measured pre- and post-chemotherapy were employed. To obtain the SE of the mean differences, the authors adhered to the following Cochrane Handbook guideline (16): SE=SD of within-participant differences between pre- and post-chemotherapy measurements divided by the square root of the number of participants. Effect sizes were derived using both the mean skeletal muscle index differences (pre- and post-chemotherapy) and the SE. Subsequently, the authors performed a meta-analysis using a random-effects model to investigate skeletal muscle index change. Between-study heterogeneity was assessed by visually inspecting forest plots generated from the study data. To identify sex-specific differences in skeletal muscle mass changes, we applied subgroup meta-analysis to explore the heterogeneity of the studies.
Results
Study characteristics and systematic review. A total of 1,048 articles, including duplicate entries in the databases, were identified as potentially relevant based on the final search terms employed. After the initial screening of titles and abstracts, we selected 15 articles for detailed analysis (Figure 1). With respect to quality assessment, all 15 of the studies reviewed were rated as high quality; their mean rating of quality assessment was 8.37, with a range from 6 to 9.
Table I shows study characteristics and the number of studies that found these characteristics to be significantly related to skeletal muscle mass loss. Table II provides more detailed information for each study reviewed. The 15 articles involved only case-control or cohort (level IV) study designs (17). Across the 15 articles, the average age of participants was 60±6.5 years. A total of 2,662 participants produced the results synthesized herein, and the mean sample size was 177±136.4, with a range from 28 to 524. Notably, 60% of the articles were published in the past 2 years (2018–2019). A large variety of cancer types were addressed in these studies; advanced non-small cell lung cancer was the most frequently reported, followed by esophageal cancer and gastric cancer equally. Among the various factors related to skeletal muscle mass change, overall survival was the factor most often identified, followed by mortality risk and sex.
General analysis of the studies reviewed (N=15).
Summary of the 15 articles reviewed.
Meta-analysis. Meta-analysis of 14 studies was performed for skeletal muscle mass change outcomes. One study did not report total skeletal muscle index values, instead it reported separate male and female skeletal muscle index results; consequently, although this study was included in the systematic review, it was not subjected to meta-analysis. Forest plots were constructed for each outcome using the mean effect size measured by the random-effects model (Figure 2).
Statistically significant heterogeneity was found (I2=86.83%, Q=132.20, p=0.00), and thus we employed a random-effects method. The summary mean difference, which was derived from the 14 studies with a total of 2,528 participants, revealed a significant loss of skeletal muscle mass (mean difference in skeletal muscle index=2.72, 95%CI=1.77-3.67, p=0.00).
Random-effects meta-analysis results for skeletal muscle mass change in patients with cancer.
In addition, four studies reported skeletal muscle index according to sex (Figure 3). Based on the mean difference by sex, statistically significant heterogeneity was found in male participants (I2=78.21%, Q=12.22, p=0.01). Skeletal muscle mass loss was found in male participants (mean difference in skeletal muscle index=4.52, 95%CI=3.34-5.71, p=0.00). In female participants, statistically significant heterogeneity was also found in the effect size (I2=89.98%, Q=34.25, p=0.00). In addition, loss of skeletal muscle mass was found (mean difference in skeletal muscle index=2.86, 95%CI=0.81-4.92, p=0.01). Notably, skeletal muscle mass loss in males was about 1.6 times higher than that found in females.
Discussion
This systematic review and meta-analysis contribute to our understanding of the current research status on changes in skeletal muscle mass during cancer treatment. Although sarcopenia has been recognized as an important issue in oncology over the past 10 years, it is only recently that studies have focused on comparing skeletal muscle index before and after cancer treatment. Skeletal muscle mass change during treatment is recognized as a predictor of chemotherapy toxicity as well as of overall survival rates. Thus, this study's contributions include identification of key factors related to muscle mass loss, quantification of muscle mass change in terms of mean differences, and identification of directions for future study.
Among the findings of this systematic review, it is interesting to note that all 15 included studies focused on skeletal muscle mass comparison during treatment were published since 2015, and 60% of them were published in the past 2 years (2018-2019). This finding indicates that cross-sectional measurements of skeletal muscle index are trending toward longitudinal measurements in order to identify patterns of skeletal muscle mass change. Consequently, it appears that oncologists and researchers are coming to recognize the clinical significance of patterns of skeletal muscle mass change, especially with regard to patient survival.
In addition, although other review articles have reported studies providing sarcopenia findings (18-20), comparative studies of skeletal muscle mass change during chemotherapy that involve skeletal muscle index measurements have been scarce. In fact, although loss of skeletal muscle mass is a measurable predictor of cancer treatment toxicity, skeletal muscle mass change during treatment has not been widely studied by race/ethnicity or by country; this is true despite the fact that the incidence of sarcopenia is known to differ by race/ethnicity (21-24). Clearly more research is needed, particularly in Western countries, to better understand the relationships of race/ethnicity and culture to skeletal muscle mass change.
Random-effects meta-analysis results for skeletal muscle mass change in patients with cancer, by sex.
Of the various cancer types addressed, advanced non-small cell lung cancer was the most frequently reported, followed by gastric cancer and esophageal cancer equally. In the studies involving advanced non-small cell lung cancer, changes in skeletal muscle index varied, ranging from 0.2 to 3.3. Similarly, a previous systematic review and meta-analysis revealed that in one study, non-small cell lung cancer showed the highest prevalence of sarcopenia at baseline (74%) (20). In addition, the study concluded that maintaining and gaining skeletal muscle cross-sectional area was significantly related to longer overall patient survival (25). In another review paper, focused on gastric cancer, (26), the authors pointed out that among patients with gastric cancer associated with eating disorders, loss of weight and muscle was often experienced, highlighting the importance of sarcopenia. Considering that chemotherapy regimens and cancer-related symptoms differ by cancer type, future researchers should compare skeletal muscle mass change among various cancer types, explore which regimens and cancer types result in the greatest skeletal muscle index loss, and identify predictors of skeletal muscle index change. These efforts may generate information that can be used to optimize interventions to better preserve and improve skeletal muscle mass.
In our systematic review, key factors related to skeletal muscle mass change included clinical, demographic, physical, and health-related factors. To be specific, overall survival was most often reported as a key factor, followed by mortality risk and sex. In addition, skeletal muscle mass loss showed associations with higher tumor stage, acute toxicities, lower physical function (in terms of hand-grip strength), chemotherapy regimens, cancer treatment type, tumor type, and tumor diameter change. While we identified a considerable variety of key factors related to skeletal muscle mass loss, we noted that cancer-related symptoms such as fatigue, depression, pain, and sleep disorder have seldom been considered in terms of their relationship with skeletal muscle mass loss. Considering that most cancer patients experience a variety of symptoms (27-29), studies examining cancer-related symptoms and associated skeletal muscle mass change should be conducted.
In the meta-analysis portion of our study, 14 studies reporting skeletal muscle mass change during cancer treatment were analyzed. No randomized controlled trials were included in this review, in order to avoid confounding factors (such as exercise and nutrition interventions) that affect muscle mass change. In our most significant finding, the mean value for skeletal muscle mass loss was 2.72 (95%CI=1.77-3.67, p=0.00, I2=86.83%); the high I2 indicates that significant heterogeneity was observed across the 14 studies. In addition, although advanced non-small cell lung cancer was the most common cancer type observed in the systematic review, the meta-analysis revealed that patients with esophageal cancer (30) experienced the greatest skeletal muscle mass loss (skeletal muscle index change: 5.6, ranging from 3.7 to 7.5) between pre- and post-neoadjuvant therapy (mean time interval: 92 days, ranging from 61 to 118 days). In addition, 89% of those cancer patients used the CROSS protocol (cisplatin/5-fluorouracil, 40 Gy/15 fr, or carboplatin/paclitaxel, 41.4 Gy/23 fr) adapted from the Dutch ChemoRadiotherapy for Oesophageal cancer followed by Surgery Study (CROSS) trial. In a previous retrospective study involving patients with esophageal cancer who underwent neoadjuvant chemoradiotherapy, patients with sarcopenia were more likely to experience severe adverse events such as fever, mucositis, and neutropenic fewer than patients without sarcopenia (31). Thus, while patients with multiple cancer types exhibited significant skeletal muscle mass loss during treatment, esophageal cancer patients undergoing neoadjuvant therapy showed twice as much skeletal muscle mass loss compared to other cancers. This may be due to the fact that eating is more challenging for esophageal cancer patients.
Among the 14 studies, 4 showed specific skeletal muscle mass change by sex. Notably, the skeletal muscle mass loss in males was about 1.6 times higher than in females. In a previous study on risk factor analysis for sarcopenia among cancer patients, males were more likely to develop sarcopenia than females (32). Possible explanations for the differences in skeletal muscle mass loss by sex include hormonal, inflammatory, and myocellular mechanisms that affect underlying biological processes that promote fat deposition and loss of lean mass and strength (33). Sex-specific hormone changes due to testosterone may be one factor that affects muscle and fat composition (34). Another possibility may be that since the male skeletal muscle index at baseline is higher than that of females (4.54 vs. 2.86), males may be more susceptible to chemotherapy's effects on skeletal muscle mass loss. Also, it is necessary to take a closer look at adherence to health-related behavioral interventions such as nutrition and exercise during cancer treatment by sex. Since only 4 of the 14 studies compared skeletal muscle index by sex, future studies examining differences in skeletal muscle index by sex during chemotherapy are needed.
This meta-analysis has some limitations that should be acknowledged. To accurately compare skeletal muscle indexes and avoid confounding factors, this meta-analysis did not include intervention studies that involved maintaining or increasing skeletal muscle index. Future systematic reviews and meta-analyses should focus on reports of nutrition and exercise interventions to better understand their effectiveness at preserving skeletal muscle mass during chemotherapy. In addition, future studies should pay attention to a number of other factors potentially contributing to skeletal muscle mass loss, including social determinants of health, recurrence status, pre-existing medical conditions, and other cancer treatments received (such as surgery, radiation therapy, hormonal therapy, or immunotherapy). Finally, more than three-quarters of the studies identified for this meta-analysis employed retrospective designs; additional prospective studies are needed to more accurately determine risk factors for skeletal muscle mass loss while also avoiding confounding factors.
In summary, the findings derived from 14 studies revealed that total skeletal muscle mass significantly declined during chemotherapy and that male patients experienced a relative muscle mass loss 1.6 times greater than female patients. Also, esophageal cancer patients undergoing chemotherapy were found to be at serious risk for skeletal muscle mass loss. Therefore, health care providers should recognize sex-specific differences in muscle mass loss and consider adjusting patients' treatment regimens accordingly.
Acknowledgements
The Authors gratefully acknowledge the assistance of Rebecca Raszewski, MS, AHIP, with the database search performed for this study. This work was supported by the National Institute of Nursing Research of the US National Institutes of Health (K24NR015340). The content is solely the responsibility of the Authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Authors' Contributions
M.K.J. designed the aim of the review, wrote the article, and supervised all steps in producing the article. C.P. contributed to the process and findings of the meta-analysis. S.H., H.L., E.R., and A.Z.D. contributed equally to this work, generated the figures, and wrote the article.
This article is freely accessible online.
Conflicts of Interest
The Authors have no conflicts of interest concerning this study.
- Received March 17, 2020.
- Revision received March 23, 2020.
- Accepted March 24, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved