Abstract
Background
Computed tomography (CT)‐defined skeletal muscle depletion and malnutrition are demonstrated as poor prognostic factors in patients with head and neck cancer (HNC), however to date, have only been explored in isolation. We aimed to describe body composition profile and examine the impact of nutritional status as well as independently and concurrently occurring body composition features on overall survival, treatment completion, unplanned admissions and length of stay (LOS) in patients undergoing radiotherapy (RT) or chemoradiotherapy (CRT) of curative intent for HNC.
Methods
This work is a retrospective, observational study of patients who had completed treatment of curative intent for HNC. Scored Patient‐Generated Subjective Global Assessment (PG‐SGA) was used to determine nutritional status. Tissue‐density data were derived at the third lumbar vertebra (L3) with sarcopenia and myosteatosis defined by published, sex‐specific threshold values stratified by body mass index for skeletal muscle index (cm2/m2) and skeletal muscle radiodensity (SMR, Hounsfield Unit).
Results
Pre‐treatment data (n = 277: 78% male, mean (SD) age 60 (13) years) revealed the prevalence of malnutrition (24.9%), sarcopenia (52.3%), myosteatosis (82.3%), and concurrently occurring sarcopenia and myosteatosis (39.7%). Malnutrition was independently associated with reduced OS for patients with moderate [hazard ratio (HR) 2.57; 95% confidence interval (CI) 1.45–4.55, P = 0.001] and severe (HR 3.19; 95% CI 1.44–7.07, P = 0.004) malnutrition on multivariable analysis but not sarcopenia (HR 1.09; 95% CI 0.70–1.71), P = 0.700 or myosteatosis (HR 1.28; 95% CI 0.57–2.84), P = 0.500). Malnutrition was associated with treatment discontinuation (P < 0.001), not completing RT as planned (P < 0.001), unplanned hospital admission (P = 0.021), and greater LOS (P < 0.001). Skeletal muscle status features were associated with unplanned hospital admissions for those with no features (32%), with sarcopenia only (50%), myosteatosis only (25%), and concurrent sarcopenia and myosteatosis (50%), P < 0.001. Similarly, a clinically relevant greater median (Q1, Q3) LOS was observed for those with sarcopenia only [5 (3, 32)], myosteatosis only [10 (5, 30)], concurrent sarcopenia, and myosteatosis [14 (4, 33)] days vs. those with no features [3 (2, 11)] days, P = 0.2.
Conclusions
Malnutrition was a more powerful prognostic indicator than CT‐defined skeletal muscle depletion and was independently associated with reduced OS in patients undergoing RT or CRT of curative intent for HNC. CT‐defined skeletal muscle depletion studies should recognize the multifaceted nature of human body composition and also measure nutritional status using validated methods in order to move towards developing a typology of high risk criteria for this complex patient group.
Keywords: Head and neck cancer, Sarcopenia, Myosteatosis, Body composition, Muscle mass, Muscle radiodensity, Radiotherapy, Malnutrition, Computed tomography, Survival
Introduction
Head and neck cancers (HNC) comprise tumours of the lip, oral cavity, oropharynx, hypopharynx, larynx, nasopharynx, and salivary glands accounting for more than 650 000 cases and 330 000 deaths annually. 1 The high prevalence of skeletal muscle depletion (6.6–70.9%) 2 and malnutrition (30–50%) 3 , 4 in patients with HNC impacts negatively on clinical, cost, and patient‐centred outcomes. The multifactorial syndrome of cancer‐associated malnutrition is characterized by progressive skeletal muscle loss with or without adipose tissue loss arising from the combination of reduced food intake and metabolic derangements. 5 , 6 , 7 The Global Leadership Initiative on Malnutrition (GLIM) recommends the diagnostic criteria for malnutrition includes at least one phenotypic [involuntary weight loss, low body mass index (BMI), or reduced muscle mass] and one etiologic (reduced food intake or assimilation, disease burden, or inflammatory condition) criterion. 8 Sarcopenia 9 and more recently, myosteatosis 10 have been recognized as independent poor prognostic factors in patients with cancer regardless of weight or nutritional status.
Sarcopenia, defined as low muscle mass, and quantified from computed tomography (CT) images as low skeletal muscle index (SMI) is a demonstrated poor prognostic factor in patients with HNC in latest meta‐analyses 11 , 12 , 13 although consensus regarding sarcopenia assessment, definitions, and reporting is still warranted. The gold standard of body composition analysis at the tissue‐organ level is at the level of the third lumbar vertebra (L3) 14 , 15 , 16 ; however, HNC studies have also reported findings from other, yet to be validated, anatomical landmarks. The availability of diagnostic positron emission tomography‐computed tomography (PET‐CT) images taken as routine care pre‐treatment and post‐treatment in some centres facilitate the evaluation of sarcopenia in patients with HNC at L3 where visceral and subcutaneous adiposity can also be visualized and annotated, an important consideration in an era of sarcopenic obesity. Our recent systematic review also demonstrated associations between sarcopenia and both radiotherapy (RT) treatment breaks and chemotherapy dose‐limiting toxicities. 2
Myosteatosis, defined as low skeletal muscle radiodensity (SMR), is a further radiologically defined prognostic marker inversely related to muscle lipid infiltration. 17 We have recently demonstrated that myosteatosis may also hold prognostic value in patients with HNC 18 which has yet to be reported elsewhere and therefore justifies further investigation. To date, studies exploring the impact of skeletal muscle depletion in patients with HNC have examined only low SMI in isolation without evaluation of concurrent low SMR. This has been suggested in other cancer populations as inadequate for description of body composition status for any given individual and that comprehensive assessment of potential combinations of body composition features is needed. 19
Low skeletal muscle mass is one potential phenotypic criterion for malnutrition; however, malnutrition itself is also a recognized poor prognostic factor, 20 but our recent systematic review determined this has not yet been explored in conjunction with skeletal muscle depletion in patients with HNC. 2 The Scored Patient‐Generated Subjective Global Assessment (PG‐SGA) is a comprehensive nutrition assessment tool validated for use in oncology populations 21 and recommended for use in routine practice by evidence‐based guidelines. 22 The tool comprises both a scored component based on patient‐generated details of weight history, food intake, nutrition‐impact symptoms and evaluation of activities and function, and also a clinician‐generated global assessment of fat stores, muscle status, and fluid status based on physical examination. This tool was not designed to detect sarcopenia or myosteatosis; however, it is consistent with the GLIM criteria and remains a valid and widely used method to diagnose malnutrition.
We aimed to describe the body composition profile of patients and examine the association between (i) nutritional status and (ii) independently or concurrently occurring body composition features comprising sarcopenia and myosteatosis on outcomes including overall survival (OS), treatment completion, unplanned admissions and length of stay (LOS) in patients undergoing adjuvant or definitive RT or chemoradiotherapy (CRT) of curative intent for HNC.
Materials and methods
Study population and study design
A retrospective, observational cohort study was undertaken in adults (≥18 years) with a primary diagnosis of HNC treated with adjuvant or definitive RT or CRT of curative intent within a 5 year period between January 2013 and December 2017. Data of consecutive patients who commenced treatment during the study period were extracted from the Department of Radiation Oncology electronic medical record (eMR) system of a single tertiary treatment centre for HNC. Eligibility criteria included the availability of pre‐treatment PET‐CT scans taken as routine care. Exclusion criteria were no PET‐CT scans available or of suitable quality for analysis and diagnoses that may occur in the head and neck region but do not fall into the accepted diagnostic grouping for HNC, for example, sarcoma.
Outcome measures
The primary outcome was OS, defined as death from any cause, where primary analysis compared OS (years) between skeletal muscle depletion (either sarcopenia or myosteatosis) or no skeletal muscle depletion and also between malnourished and well‐nourished groups. OS was calculated from the end of RT or the last known date alive. Secondary outcomes included treatment completion, RT and CRT prescribed versus delivered, unplanned hospital admissions, and LOS (days).
Nutritional status and intervention
Nutritional status was assessed using the PG‐SGA 21 in line with best practice in oncology populations, evidence‐based guidelines for nutritional management of patients with HNC 22 and consistent with GLIM malnutrition diagnostic criteria. The PG‐SGA categorizes patients as well‐nourished (A) or either moderately (B) or severely (C) malnourished. Patients received nutrition assessment and intervention according to standard care derived from evidence‐based guidelines 22 with improved uptake achieved through a prior implementation science approach. 23 In our centre, nutrition intervention commonly involves nutrition support delivered via prophylactic gastrostomy in patients receiving multimodal treatment, particularly for those requiring adjuvant or definitive CRT. Nasogastric tube feeding was commenced for patients without a gastrostomy who were unable to maintain adequate nutrition and hydration orally. Enteral or oral nutrition support was initiated when clinically indicated according to nutritional status, weight, percentage weight loss, and nutritional intake. Routine care for high nutrition‐risk patients in the treating centre aims for dietetic assessment weekly during treatment and for as long as needed in the post‐treatment recovery and rehabilitation phase until weight and nutritional intake are stabilized and tube feeding is no longer clinically indicated.
Data collection
Patient characteristics, diagnosis, and treatment demographics were collected from the eMR and paper‐based hospital records with death data current to 23 June 2020 obtained from the New South Wales Registry of Births, Deaths and Marriages. 24 Ethnicity (self‐reported) was captured from the eMR and classified according to the Australian Standard Classification of Cultural and Ethnic Groups. 25 All research data was maintained using Sydney Local Health District hosted Research Electronic Data Capture. 26
Body composition analysis by computed tomography
Cross‐sectional tissue‐density data were derived from the CT component of PET‐CT images taken for routine cancer staging at diagnosis. An individual, trained observer (MF) blind to patients' outcomes analysed single axial images at L3 using Slice‐O‐Matic, Version 5.0 (Tomovision, Magog, QC, Canada). All images were 3 mm thickness with a peak kilovoltage of 120. Recognized Hounsfield [Hounsfield unit (HU)] threshold values were applied for muscle (−29 to +150 HU), visceral adipose tissue (VAT, −50 to −150 HU), and subcutaneous adipose tissue (SAT, −30 to −190 HU). Adipose tissue external to the abdominal wall but within the muscle fascia was annotated as SAT with both SAT and VAT summed to yield total adipose tissue (TAT) as described by Mishra et al. 27 Mean surface area (cm2) for skeletal muscle, VAT, SAT and TAT and mean (HU) SMR were normalized for height (m2) and reported as SMI (cm2/m2), visceral adipose tissue index (VATI, cm2/m2), subcutaneous adipose tissue index (cm2/m2), and total adipose tissue index (cm2/m2). Sex‐specific, BMI‐stratified threshold values published by Martin et al. were applied for classifying patients with sarcopenia and myosteatosis. 28 Sarcopenia was defined as an SMI for female patients < 41 cm2/m2 and for male patients < 43 cm2/m2 (BMI < 25.0 kg/m2) and <53 cm2/m2 (BMI ≥ 25 kg/m2). Myosteatosis was defined for both sexes as a mean SMR < 41 HU (BMI < 25.0 kg/m2) and 33 HU (BMI ≥ 25 kg/m2). The number and combination of body composition features exhibited by individuals was also captured to explore whether skeletal muscle depletion and obesity (defined as BMI ≥ 30 kg/m2) influence outcomes when occurring in isolation or concurrently.
Statistical analysis
Differences were analysed using paired t‐tests and Pearson χ 2 test for continuous and categorical variables, respectively. Non‐parametric variables were analysed using the Kruskal–Wallis test. Data are presented as mean (SD) and median (range or interquartile range) according to distribution normality. Survival curves were constructed using the Kaplan–Meier method with log‐rank test reported. Unadjusted and adjusted Cox proportional hazard regression analyses were performed to determine hazard ratios (HR) that are reported with corresponding 95% confidence intervals (CIs). Covariates were investigated according to their prognostic value and included age, sex, ethnicity, disease stage, performance status, Charlson comorbidity index (CCI), BMI, human papilloma virus (HPV) status, alcohol use, and smoking history. Analysis was completed using R Version 1.2.5042 (Vienna, Austria). 29 Statistical significance was considered at the level of P < 0.05 with all results also evaluated for clinical importance and implications for practice.
Ethics approval and reporting
The study was approved by the Human Research Ethics Committee at Royal Prince Alfred Hospital, Sydney, Australia (HREC/14/RPAH/524) with site‐specific approval for this study to be conducted across Royal Prince Alfred Hospital (SSA/15/RPAH/148) and Chris O'Brien Lifehouse (LH15.017). The study is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology statement to ensure comprehensive reporting.
Results
Within the study period, 359 patients underwent adjuvant or definitive RT or CRT for HNC. Of these, 277 patients had evaluable scans and met the inclusion criteria. Patients without evaluable scans were slightly older with mean (SD) age of 65 (13) vs. 60 years (P = 0.003), female (40% vs. 22%, P = 0.002 and more likely to have tumours of the oral cavity (P < 0.001). Scan availability can be dependent upon whether the staging investigations were undertaken in a centre elsewhere and also variation in clinician practices with regards to requesting PET‐CT imaging for specific diagnoses.
Patient characteristics
Baseline demographics and clinical characteristics of included patients (n = 277) are presented in Table 1. The majority were male (78%), tumour type was predominantly squamous cell carcinoma (85%) with the three most common tumour sites oropharynx (41%), nasopharynx (15%), and larynx (13%). Most patients (79%) had advanced clinical stage (III/IV) according to American Joint Committee on Cancer 7th Edition 30 requiring multimodal treatment regimens (86%). Baseline nutrition and body composition characteristics are presented in Table 2. Body composition features differed by sex including SMI, SMR, VATI, SATI and TATI.
Table 1.
Baseline demographics and clinical characteristics by skeletal muscle features of sarcopenia defined by low SMI and myosteatosis defined by low SMR
| Overall, N (%) | SMI, N (%) | SMR, N (%) | |||||
|---|---|---|---|---|---|---|---|
| Characteristic | Total (N = 277) | Normal (N = 132 a ) | Low (N = 145 a ) | P value b | Normal (N = 49 a ) | Low (N = 228 a ) | P value b |
| Age at diagnosis, mean (SD), years | 60 (13) | 58 (13) | 62 (13) | 0.005 | 49 (12) | 62 (12) | <0.001 |
| Gender | 0.65 | 0.10 | |||||
| Male | 216 (78%) | 105 (80%) | 111 (77%) | 43 (88%) | 173 (76%) | ||
| Female | 61 (22%) | 27 (20%) | 34 (23%) | 6 (12%) | 55 (24%) | ||
| Ethnicity | 0.14 | 0.003 | |||||
| Asian | 58 (21%) | 22 (17%) | 36 (25%) | 19 (39%) | 39 (17%) | ||
| European | 51 (18%) | 23 (17%) | 28 (19%) | 3 (6.1%) | 48 (21%) | ||
| Oceanian | 152 (55%) | 76 (58%) | 76 (52%) | 24 (49%) | 128 (56%) | ||
| Other | 16 (5.8%) | 11 (8.3%) | 5 (3.4%) | 3 (6.1%) | 13 (5.7%) | ||
| Performance status | 0.043 | 0.018 | |||||
| ECOG 0 | 149 (54%) | 82 (62%) | 67 (46%) | 37 (76%) | 112 (49%) | ||
| ECOG 1 | 85 (31%) | 38 (29%) | 47 (32%) | 11 (22%) | 74 (32%) | ||
| ECOG 2 | 19 (6.9%) | 5 (3.8%) | 14 (9.7%) | 1 (2.0%) | 18 (7.9%) | ||
| ECOG 3 | 2 (0.7%) | 0 (0%) | 2 (1.4%) | 0 (0%) | 2 (0.9%) | ||
| ECOG 4 | 1 (0.4%) | 0 (0%) | 1 (0.7%) | 0 (0%) | 1 (0.4%) | ||
| Not documented | 21 (7.6%) | 7 (5.3%) | 14 (9.7%) | 0 (0%) | 21 (9.2%) | ||
| CCI, total, median (range) | 5 (1–15) | 4 (2–13) | 5 (1–15) | 0.021 | 3 (1–15) | 5 (2–14) | <0.001 |
| CCI | 0.10 | <0.001 | |||||
| ≥6 | 116 (42%) | 48 (36%) | 68 (47%) | 8 (16%) | 108 (47%) | ||
| <6 | 161 (58%) | 84 (64%) | 77 (53%) | 41 (84%) | 120 (53%) | ||
| Disease stage (AJCC7) | 0.86 | 0.41 | |||||
| I | 10 (3.6%) | 6 (4.5%) | 4 (2.8%) | 4 (8.2%) | 6 (2.6%) | ||
| II | 25 (9.0%) | 12 (9.1%) | 13 (9.0%) | 4 (8.2%) | 21 (9.2%) | ||
| III | 50 (18%) | 22 (17%) | 28 (19%) | 9 (18%) | 41 (18%) | ||
| IV | 168 (61%) | 79 (60%) | 89 (61%) | 27 (55%) | 141 (62%) | ||
| Other | 24 (8.7%) | 13 (9.8%) | 11 (7.6%) | 5 (10%) | 19 (8.3%) | ||
| Treatment modality | 0.86 | 0.70 | |||||
| RT—definitive | 38 (14%) | 16 (12%) | 22 (15%) | 7 (14%) | 31 (14%) | ||
| CRT—definitive | 132 (48%) | 65 (49%) | 67 (46%) | 21 (43%) | 111 (49%) | ||
| Surgery + CRT—adjuvant | 42 (15%) | 21 (16%) | 21 (14%) | 10 (20%) | 32 (14%) | ||
| Surgery + RT—adjuvant | 65 (23%) | 30 (23%) | 35 (24%) | 11 (22%) | 54 (24%) | ||
| Tumour site | 0.44 | 0.019 | |||||
| Oral cavity/lip | 28 (10%) | 13 (9.8%) | 15 (10%) | 5 (10%) | 23 (10%) | ||
| Oropharynx | 113 (41%) | 54 (41%) | 59 (41%) | 17 (35%) | 96 (42%) | ||
| Hypopharynx | 14 (5.1%) | 3 (2.3%) | 11 (7.6%) | 1 (2.0%) | 13 (5.7%) | ||
| Larynx | 37 (13%) | 22 (17%) | 15 (10%) | 2 (4.1%) | 35 (15%) | ||
| Nasopharynx | 41 (15%) | 20 (15%) | 21 (14%) | 15 (31%) | 26 (11%) | ||
| Salivary gland | 20 (7.2%) | 8 (6.1%) | 12 (8.3%) | 6 (12%) | 14 (6.1%) | ||
| Nasal cavity/Paranasal sinus | 11 (4.0%) | 6 (4.5%) | 5 (3.4%) | 1 (2.0%) | 10 (4.4%) | ||
| Other | 12 (4.3%) | 5 (3.8%) | 7 (4.8%) | 2 (4.1%) | 10 (4.4%) | ||
| Unknown primary | 1 (0.4%) | 1 (0.8%) | 0 (0%) | 0 (0%) | 1 (0.4%) | ||
| Tumour type | 0.31 | 0.001 | |||||
| Squamous cell carcinoma | 236 (85%) | 108 (82%) | 128 (88%) | 34 (69%) | 202 (89%) | ||
| Other | 38 (14%) | 22 (17%) | 16 (11%) | 13 (27%) | 25 (11%) | ||
| Not documented | 3 (1.1%) | 2 (1.5%) | 1 (0.7%) | 2 (4.1%) | 1 (0.4%) | ||
| Smoking status | 0.21 | 0.003 | |||||
| Never smoked | 97 (35%) | 51 (39%) | 46 (32%) | 25 (51%) | 72 (32%) | ||
| Current smoker | 56 (20%) | 20 (15%) | 36 (25%) | 5 (10%) | 51 (22%) | ||
| Previous smoker | 112 (40%) | 56 (42%) | 56 (39%) | 14 (29%) | 98 (43%) | ||
| Not documented | 12 (4.3%) | 5 (3.8%) | 7 (4.8%) | 5 (10%) | 7 (3.1%) | ||
| Alcohol use | 0.36 | 0.16 | |||||
| None or social only | 157 (57%) | 82 (62%) | 75 (52%) | 33 (67%) | 124 (54%) | ||
| 1–2 standard drinks/day | 26 (9.4%) | 11 (8.3%) | 15 (10%) | 4 (8.2%) | 22 (9.6%) | ||
| >2 standard drinks/day | 67 (24%) | 27 (20%) | 40 (28%) | 6 (12%) | 61 (27%) | ||
| Not documented | 27 (9.7%) | 12 (9.1%) | 15 (10%) | 6 (12%) | 21 (9.2%) | ||
| HPV status | 0.70 | 0.28 | |||||
| Negative | 27 (9.7%) | 10 (7.6%) | 17 (12%) | 2 (4.1%) | 25 (11%) | ||
| Positive | 84 (30%) | 42 (32%) | 42 (29%) | 14 (29%) | 70 (31%) | ||
| Not applicable | 164 (59%) | 79 (60%) | 85 (59%) | 32 (65%) | 132 (58%) | ||
| Not documented | 2 (0.7%) | 1 (0.8%) | 1 (0.7%) | 1 (2.0%) | 1 (0.4%) | ||
AJCC7, American Joint Committee on Cancer, 7th Edition; CCI, Charlson comorbidity index; CRT, chemoradiotherapy; HPV, human papilloma virus; RT, radiotherapy; SMI, skeletal muscle index; SMR, skeletal muscle radiodensity.
Statistics presented: median (minimum–maximum); mean (SD); n (%).
Statistical tests performed: t‐test; χ 2 test of independence.
Table 2.
Baseline nutrition and body composition characteristics by skeletal muscle features
| Overall | SMI | SMR | ||||||
|---|---|---|---|---|---|---|---|---|
| Characteristic | (N = 277; M:216; F:61) | Normal (N = 132 a ; M:105; F:27) | Low (N = 145 a ; M:111; F:34) | P value b | Normal (N = 49 a ; M:43; F:6) | Low (N = 228 a ; M:173; F:55) | P value b | |
| Height (cm), mean (SD) | 170 (9) | 170 (9) | 170 (9) | 0.91 | 174 (7) | 169 (9) | <0.001 | |
| Weight (kg), mean (SD) | 75 (18) | 80 (18) | 70 (17) | <0.001 | 76 (15) | 75 (19) | 0.55 | |
| BMI (kg/m2) mean, (SD) | 25.8 (5.4) | 27.6 (5.4) | 24.1 (4.9) | <0.001 | 25.2 (4.4) | 26.0 (5.6) | 0.28 | |
| BMI category, (kg/m2), N (%) | <0.001 | <0.001 | ||||||
| <20.0 | 37 (13%) | 3 (2.3%) | 34 (23%) | 8 (16%) | 29 (13%) | |||
| 20.0 to 24.9 | 94 (34%) | 50 (38%) | 44 (30%) | 7 (14%) | 87 (38%) | |||
| 25.0 to 29.9 | 88 (32%) | 38 (29%) | 50 (34%) | 29 (59%) | 59 (26%) | |||
| ≥30 | 58 (21%) | 41 (31%) | 17 (12%) | 5 (10%) | 53 (23%) | |||
| SMA (cm2), mean (SD) | ||||||||
| Males | 146 (32) | 166 (27) | 128 (24) | <0.001 | 168 (29) | 141 (31) | <0.001 | |
| Females | 105 (22) | 123 (18) | 90 (11) | <0.001 | 100 (15) | 105 (23) | 0.47 | |
| SMI (cm2/m2), mean (SD) | ||||||||
| Males | 49 (10) | 56 (8) | 43 (7) | <0.001 | 54 (8) | 48 (9) | <0.001 | |
| Females | 40 (8) | 47 (5) | 35 (4) | <0.001 | 39 (5) | 40 (8) | 0.50 | |
| SMR (HU), mean (SD) | ||||||||
| Males | 30 (7) | 32 (6) | 28 (7) | <0.001 | 38 (4) | 28 (6) | <0.001 | |
| Females | 29 (8) | 29 (7) | 29 (9) | 0.84 | 41 (5) | 28 (7) | <0.001 | |
| VATI (cm2/m2), median (range) | ||||||||
| Males | 48 (1–180) | 50 (3–180) | 46 (1–154) | 0.57 | 38 (1–116) | 50 (2–180) | <0.001 | |
| Females | 31 (3–113) | 33 (2–113) | 31 (3–76) | 0.038 | 5 (3–31) | 34 (3–113) | 0.002 | |
| SATI (cm2/m2), median (range) | ||||||||
| Males | 48 (1–221) | 51 (3–221) | 45 (1–130) | 0.011 | 46 (1–123) | 48 (3–221) | 0.15 | |
| Females | 77 (14–221) | 96 (18–221) | 61 (14–193) | 0.011 | 32 (14–125) | 78 (18–221) | 0.13 | |
| TATI (cm2/m2), median (range) | ||||||||
| Males | 97 (2–353) | 97 (10–353) | 97 (2–242) | 0.087 | 91 (2–228) | 102 (6–353) | 0.006 | |
| Females | 116 (17–277) | 141 (21–277) | 94 (17–242) | 0.007 | 35 (17–156) | 117 (21–277) | 0.050 | |
| Nutritional status, PG‐SGA score, median (range) | ||||||||
| 5 (1–42) | 3 (1–25) | 7 (1–42) | <0.001 | 4 (1–42) | 5 (1–40) | 0.45 | ||
| Nutritional status, PG‐SGA category, N (%) | 0.004 | 0.053 | ||||||
| A (well‐nourished) | 175 (63%) | 97 (73%) | 78 (54%) | 39 (80%) | 136 (60%) | |||
| B (moderately malnourished) | 50 (18%) | 18 (14%) | 32 (22%) | 5 (10%) | 45 (20%) | |||
| C (severely malnourished) | 19 (6.9%) | 4 (3.0%) | 15 (10%) | 3 (6.1%) | 16 (7.0%) | |||
| Not documented | 33 (12%) | 13 (9.8%) | 20 (14%) | 2 (4.1%) | 31 (14%) | |||
| Nutrition support, N (%) | 0.22 | 0.87 | ||||||
| NGT | 32 (18%) | 10 (13%) | 22 (23%) | 5 (17%) | 27 (19%) | |||
| Gastrostomy—PEG | 121 (70%) | 58 (75%) | 63 (65%) | 22 (76%) | 99 (68%) | |||
| Gastrostomy—RIG | 18 (10%) | 8 (10%) | 10 (10%) | 2 (6.9%) | 16 (11%) | |||
| Gastrostomy—surgical | 2 (1.1%) | 0 (0%) | 2 (2.1%) | 0 (0%) | 2 (1.4%) | |||
| TPN | 1 (0.6%) | 1 (1.3%) | 0 (0%) | 0 (0%) | 1 (0.7%) | |||
BMI, body mass index; HU, Hounsfield units; NGT, nasogastric tube; PEG, percutaneous endoscopic gastrostomy; PG‐SGA, Patient‐Generated Subjective Global Assessment; RIG, radiologically inserted gastrostomy; SATI, subcutaneous adipose tissue index; SMA, skeletal muscle area; SMI, skeletal muscle index; SMR, skeletal muscle radiodensity; TATI, total adipose tissue index; TPN, total parenteral nutrition; VATI, visceral adipose tissue index.
Statistics presented: median (minimum–maximum); mean (SD); n (%).
Statistical tests performed: t‐test; χ 2 test of independence.
Skeletal muscle status, combination of body composition features, and nutritional status
The prevalence of the various combinations of body composition features in conjunction with nutritional status as determined by the PG‐SGA Global Category is shown in Figure 1. Malnutrition was prevalent in 24.9% of patients, and while most frequently occurring in those with concurrent sarcopenia and myosteatosis (14.4%), it appeared to be present irrespective of body composition features. Sarcopenia prevalence overall was 52.3%; however, it was only present in isolation in a small proportion of patients (6.5%), more frequently occurring concurrently with myosteatosis (39.7%) or with myosteatosis and obesity combined (6.2%). Sarcopenic obesity was not present in any patients unless myosteatosis was also present. Similarly, the overall prevalence of myosteatosis was 82.3%, which was present in isolation (23.5%) but more so concurrently with sarcopenia (39.7%), concurrently with obesity (13.0%) and with sarcopenia and obesity combined (6.1%). Only 9.4% patients did not exhibit either a single or combination of body composition features, the majority of whom were well‐nourished.
Figure 1.
Euler diagram denoting pre‐treatment combination of computed tomography‐defined body composition features and baseline nutritional status as determined by Scored Patient‐Generated Subjective Global Assessment (n = 277).
Skeletal muscle status, nutritional status, and overall survival
Associations between skeletal muscle status and OS are presented in Figure 2. The number of body composition features was associated with reduced OS (log‐rank P = 0.017) as was the combination of skeletal muscle features (log‐rank P = 0.0028). Associations between nutritional status and OS are presented in Figure 3. Nutritional status as determined by PG‐SGA Global Category was associated with reduced OS (log‐rank P < 0.0001) as was malnutrition (moderately and severely malnourished combined) (log‐rank P < 0.0001). Unadjusted and adjusted analyses are presented in Table 3. Covariates not associated with reduced OS in the cohort were excluded from the adjusted models. On adjusted analysis pre‐treatment nutritional status remained an independent predictor of reduced OS for patients who were moderately malnourished (HR = 2.57; 95% CI 1.45–4.55), P = 0.001 or severely malnourished (HR = 3.19; 95% CI 1.44–7.07), P = 0.004 as did overall malnutrition (HR = 3.03; 95% CI 1.87–4.93), P < 0.001. Sarcopenia, myosteatosis, number of body composition features, and combination of body composition features were also significant predictors of reduced OS on unadjusted analysis but did not remain so when adjusted for covariates that included malnutrition.
Figure 2.
Kaplan–Meier survival estimates of overall survival with log‐rank comparisons for: (A) number of body composition features; (B) combination of skeletal muscle status features.
Figure 3.
Kaplan–Meier survival estimates of overall survival with log‐rank comparisons for: (A) nutritional status – Patient‐Generated Subjective Global Assessment – global category; (B) nutritional status – well‐nourished vs. malnourished.
Table 3.
Regression models for associations between skeletal muscle status and nutritional status and overall survival
| Variable | Pre‐treatment | ||||
|---|---|---|---|---|---|
| Unadjusted | Adjusted a | ||||
| n | HR (95% CI) | P value | HR (95% CI) | P value | |
| Nutritional status—PG‐SGA category | |||||
| Well‐nourished (A) | 175 | 1.0 (ref)M1 | <0.001 | 1.0 (ref)A1 | <0.001 b |
| Moderately malnourished (B) | 50 | 3.14 (1.88–5.24) | 2.57 (1.45–4.55) | 0.001 b | |
| Severely malnourished (C) | 19 | 3.97 (2.02–7.82) | 3.19 (1.44–7.07) | 0.004 b | |
| Well‐nourished (A) | 175 | 1.0 (ref)M2 | <0.001 | 1.0 (ref)A2 | <0.001 b |
| Malnourished c (B and C combined) | 69 | 3.82 (2.52–5.79) | 3.03 (1.87–4.93) | ||
| Skeletal muscle status | |||||
| No sarcopenia | 132 | 1.0 (ref)M3 | 0.060 | 1.0 (ref)A3 | 0.700 |
| Sarcopenia | 145 | 1.48 (0.98–2.24) | 1.09 (0.70–1.71) | ||
| No myosteatosis | 49 | 1.0 (ref)M4 | 0.006 | 1.0 (ref) A2 | 0.500 |
| Myosteatosis | 228 | 2.75 (1.33–5.68) | 1.28 (0.57–2.84) | ||
| Body composition features—number | |||||
| No features | 26 | 1.0 (ref)M5 | 0.013 | 1.0 (ref)A4 | 0.8 |
| 1 Feature | 88 | 1.36 (0.52–6.62) | 0.83 (0.30–2.27) | ||
| 2 Features | 146 | 2.61 (1.05–6.49) | 0.96 (0.35–2.63) | ||
| 3 Features | 17 | 2.02 (0.62–6.62) | 0.92 (0.26–3.24) | ||
| Body composition features—combination | |||||
| No features | 26 | 1.0 (ref)M6 | 0.005 | 1.0 (ref)A5 | <0.001 |
| Sarcopenia only | 18 | 0.27 (0.03–2.31) | 0.26 (0.03–2.22) | 0.200 | |
| Myosteatosis only | 65 | 1.66 (0.62–4.44) | 0.94 (0.33–2.63) | >0.900 | |
| Obesity only | 5 | 2.09 (0.41–10.8) | 1.29 (0.25–2.50) | 0.800 | |
| Sarcopenia + Myosteatosis | 110 | 2.88 (1.15–7.22) | 1.07 (0.39–2.95) | 0.900 | |
| Sarcopenia + Obesity | 0 | Not defined | Not defined | ||
| Myosteatosis + Obesity | 36 | 1.88 (0.66–5.34) | 0.79 (0.27–3.39) | 0.700 | |
| Sarcopenia + Myosteatosis + Obesity | 17 | 2.02 (0.62–6.62) | 0.96 (0.27–3.39) | >0.900 | |
Models: Unadjusted (M1 to M6) and adjusted (A1 to A5).
CI, confidence interval; HR, hazard ratio; PG‐SGA, Patient‐Generated Subjective Global Assessment.
Adjusted for age, sex, ethnicity, performance status, disease stage, Charlson comorbidity index, and nutritional status (Malnourished).
Also adjusted for low muscle attenuation but not for nutritional status.
Malnourished = PG‐SGA B (moderate) or C (severe).
Treatment completion
Associations between nutritional status and skeletal muscle status vs. treatment completion are presented in Table 4. Nutritional status was associated with treatment discontinuation (P < 0.001) and whether RT was delivered as planned (<0.001). Treatment was discontinued for 10% of patients with concurrent sarcopenia and myosteatosis vs. 0% for those with no skeletal muscle status features (P = 0.14). There was no association detected between skeletal muscle status and either RT or chemotherapy completion.
Table 4.
Associations between skeletal muscle status and nutritional status vs. treatment and hospital admission outcomes
| Outcome | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Treatment discontinued | Radiotherapy delivered as planned | Systemic therapy delivered as planned a | Unplanned admission | Length of stay (days) | ||||||||
| Variable | n | n (%) b | P value c | n (%) b | P value c | n (%) b | P value c | n (%) b | P value c | Median b | (Q1, Q3) | P value d |
| Nutritional status—PG‐SGA category | ||||||||||||
| Well‐nourished (A) | 175 | 5 (2.9%) | <0.001 | 168 (96%) | <0.001 | 90 (73%) | 0.6 | 59 (34%) | 0.021 | 7 | (3, 20) | <0.001 |
| Moderately malnourished (B) | 50 | 7 (14%) | 44 (88%) | 24 (83%) | 19 (38%) | 12 | (5, 37) | |||||
| Severely malnourished (C) | 19 | 1 (5.3%) | 18 (95%) | 7 (88%) | 11 (58%) | 26 | (9, 44) | |||||
| Not documented | 33 | 7 (21%) | 23 (70%) | 10 (77%) | 19 (58%) | 40 | (14, 46) | |||||
| Skeletal muscle status | ||||||||||||
| No sarcopenia | 132 | 7 (5.3%) | 0.3 | 124 (94%) | 0.2 | 63 (73%) | 0.7 | 35 (27%) | <0.001 | 10 | (3, 26) | 0.2 |
| Sarcopenia | 145 | 13 (9.0%) | 129 (89%) | 68 (77%) | 73 (50%) | 14 | (4,34) | |||||
| No myosteatosis | 49 | 0 (0%) | 0.065 | 48 (98%) | 0.12 | 19 (61%) | 0.078 | 19 (39%) | >0.9 | 3 | (3, 20) | 0.047 |
| Myosteatosis | 228 | 20 (8.8%) | 205 (90%) | 112 (78%) | 89 (39%) | 14 | (4, 33) | |||||
| Skeletal muscle status—combination of features | ||||||||||||
| No features | 31 | 0 (0%) | 0.14 | 31 (100%) | 0.2 | 11 (55%) | 0.2 | 10 (32%) | <0.001 | 3 | (2, 11) | 0.2 |
| Sarcopenia only | 18 | 0 (0%) | 17 (94%) | 8 (73%) | 9 (50%) | 5 | (3, 32) | |||||
| Myosteatosis only | 101 | 7 (6.9%) | 93 (92%) | 52 (79%) | 25 (25%) | 10 | (5, 30) | |||||
| Sarcopenia + Myosteatosis | 127 | 13 (10%) | 112 (88%) | 60 (78%) | 64 (50%) | 14 | (4, 33) | |||||
PG‐SGA, Patient‐Generated Subjective Global Assessment.
n = 174 patients had systemic therapy.
Statistics presented: n (%); median (IQR).
Statistical tests performed: χ 2 test of independence.
Kruskal–Wallis test.
Unplanned admission and length of stay
Malnourished patients were more likely to require unplanned hospital admission with 58% of severely malnourished patients vs. 34% of well‐nourished patients admitted (P = 0.021), (Table 4). Similarly, median (Q1, Q3) LOS was greater for severely malnourished [26 (9, 44)] and moderately malnourished [12 (5, 37)] vs. well‐nourished [7 (3, 20)] days, P < 0.001. Skeletal muscle status was associated with unplanned admissions for those with sarcopenia alone (50%), myosteatosis alone (25%), or concurrent sarcopenia and myosteatosis (50%) and those with no features (32%), P < 0.001. Clinically relevant differences in LOS was also associated with skeletal muscle status with median (Q1, Q3) LOS for those with sarcopenia alone [5 (3, 32)], myosteatosis alone [10 (5, 30)], and concurrent sarcopenia and myosteatosis [14 (4, 33) vs. those with no skeletal muscle status features of 3 (2, 11)] days, P = 0.2.
Discussion
To the best of our knowledge, this is the first study to explore CT‐defined body composition profiles, and in particular, the association between (i) nutritional status and (ii) skeletal muscle status on outcomes for patients undergoing RT or CRT of curative intent for HNC. Key findings in our study were malnutrition is a more powerful prognostic factor than skeletal muscle status and is independently associated with reduced OS in patients with HNC. Importantly, malnutrition may be present, irrespective of body composition features and was also associated with treatment discontinuation, RT not completed as planned, unplanned admissions and greater length of stay. Similarly, concurrently occurring sarcopenia and myosteatosis were also associated with unplanned admission and greater length of stay. This aligns with our previous work regarding the association between sarcopenia and myosteatosis on reduced OS 18 and also broadens the existing evidence through exploration of individually or concurrently occurring body composition features in conjunction with nutritional status. Low skeletal muscle mass is a phenotypic GLIM criteriona for malnutrition diagnosis, which may also arise due to involuntary weight loss or low BMI. 8 Our study highlights the importance of understanding the complexities of human body composition and the potential impact of concurrent malnutrition on outcomes for high risk patients.
Nutritional status was independently associated with reduced OS for patients with moderate (HR 2.57; 95% CI 1.45–4.55, P = 0.001) and severe (HR 3.19; 95% CI 1.44–7.07, P = 0.004) malnutrition. Malnutrition was significantly associated with other poor outcomes including treatment discontinuation and not completing RT as planned. Malnourished patients were also more likely to require unplanned hospital admission and nutritional status significantly influenced LOS. The median (Q1, Q3) LOS for moderately malnourished [12 (5, 37)] and severely malnourished [26 (9, 44] patients vs. those who were well‐nourished [7 (3, 20)] days, P < 0.001. Malnutrition is a well‐documented high cost diagnosis, both to patients and the healthcare system. We have previously published findings regarding economic implications of unplanned hospital admissions in patients with HNC 23 and specifically those with CT‐defined sarcopenia 18 with mean (SD) unplanned admission costs reported from an organizational perspective in Australian Dollars ($AUD) of $AUD15 846 ($AUD17 707) for patients without sarcopenia vs. $AUD47 945 ($AUD82 688) for those with sarcopenia. The importance of establishing processes to facilitate early identification, intervention, and monitoring to ameliorate the detrimental sequelae of both malnutrition and skeletal muscle depletion as a contributing criterion holds potential to optimize both clinical and economic outcomes and has recently been articulated in the Clinical Oncology Society of Australia Position Statement on Cancer‐Related Malnutrition and Sarcopenia. 31
Although the association between sarcopenia and OS was not significant on adjusted analysis (HR 1.09; 95% CI 0.70–1.71), P = 0.700), our recent meta‐analysis of pooled results from seven studies (1059 patients) showed pre‐treatment L3 CT‐defined sarcopenia was independently associated with reduced OS (HR 2.07; 95% CI 1.47–2.92, P < 0.0001, I 2 = 49%). 11 This difference may be attributable, in part, to variation in results at the individual study level, which is improved on meta‐analysis. Further, to our knowledge, other studies have not reported nutritional status using a validated nutrition assessment tool such as the PG‐SGA nor subsequently controlled for this in their findings. The association between myosteatosis and OS on adjusted analysis (HR 1.28; 95% CI 0.57–2.84), P = 0.500) differs from our previous study (n = 79) (HR 8.86; 95% CI 1.12–69.88, P = 0.038) 18 which is likely attributable to the smaller cohort in the earlier study. Myosteatosis is a demonstrated poor prognostic factor in other tumour groups 32 , 33 ; however, there remains a paucity of studies in patients with HNC.
Most research on CT‐defined body composition in patients with HNC has to date reported findings regarding sarcopenia alone; however, our study explores whether skeletal muscle depletion features occur in isolation or concurrently appears to also influence outcomes. Patients with no features were less likely to require unplanned hospital admission (0%) than those with sarcopenia alone (50%), myosteatosis only (25%), concurrent sarcopenia and myosteatosis (50%), P < 0.001. Similarly, a clinically relevant difference in median (Q1, Q3) LOS was observed for those with sarcopenia only [5 (3, 32)], myosteatosis only [10 (5, 30)], concurrent sarcopenia and myosteatosis [10 (4, 33)] vs. those with no features [3 (2, 11)] days, P = 0.2. This harmonizes with findings of Martin et al. 19 in patients undergoing elective surgery for colorectal cancer, which demonstrated that CT‐defined multidimensional body habitus was independently associated with hospital admission and LOS. Xiao et al (2020) also identified even higher risk of overall mortality where sarcopenia and myosteatosis occurred concurrently in colon cancer surgical patients. 34 Human body composition is multifaceted, and therefore, reducing an individual feature to a binary proposition is an oversimplification of what are complex body habitus parameters. 35
In our study, validated methods in line with the GLIM criteria were used to diagnose both malnutrition and skeletal muscle depletion; however, no single method currently exists to detect both. With the epidemiological shift towards obesity in Western societies, both may well be occult diagnoses, with high risk patients going undetected or unrecognized. Early identification of high risk patients is vital to improving outcomes; however, clinical utility in a practical, real‐world setting must also be given consideration. CT‐defined body composition analysis requires considerable financial and resource investment in software, analyst training, time to locate scans, isolate the CT DICOM file of PET‐CT scans, and undertake the necessary landmarking and analysis which in our study is estimated to be approximately 1 h per scan per patient even with auto‐segmentation software functionality that still requires quality assurance checks. Whereas nutrition assessment with a validated assessment tool, the PG‐SGA, as recommended in evidence‐based guidelines, 22 can be undertaken as part of routine clinical practice in approximately 15 min. Further, availability of scans for analysis is beholden to diagnostic and staging imaging protocols while nutritional status can be reassessed at regular intervals as part of best‐practice nutrition monitoring with minimal resource utilization. In the quest to optimize patient care and outcomes, adhering to the fundamentals of nutrition care and clinical practice for assessment of nutritional status may also offer a timely and efficient mechanism for flagging and monitoring high risk patients.
Study strengths include the first to employ validated methods for evaluation of both nutritional status that align with the GLIM criteria and the gold standard of skeletal muscle evaluation at the tissue‐organ level. Limitations of our study include its retrospective nature and the availability of nutritional status at baseline only. Given patients with HNC are known to often present as well‐nourished at baseline but experience significant decline in nutritional status throughout the course of care, addressing this through prospective study design in a larger cohort to capture the nature of nutritional status and skeletal muscle status inter‐relationships and change over time would be of value for future research. Given malnutrition and skeletal muscle depletion are related, we acknowledge that a larger sample size may provide clarity regarding the inter‐relationships of these prognostic factors. Future prospective studies that explore the alignment between skeletal muscle depletion and malnutrition using validated methods are recommended. Overall, research in the field of CT‐defined body composition would benefit from agreed definitions of sarcopenia and myosteatosis and standardized reporting criteria to ultimately improve the certainty and applicability of the growing evidence base.
Conclusions
In this study, malnutrition was a more powerful prognostic indicator than CT‐defined skeletal muscle depletion and was independently associated with reduced OS in patients undergoing RT or CRT of curative intent for HNC. It was also associated with poor outcomes including treatment discontinuation and RT not completed as planned. Malnutrition and also the combination of skeletal muscle status features were associated with unplanned hospital admissions and LOS. Studies of CT‐defined skeletal muscle depletion should recognize the multifaceted nature of human body composition; and as such, the inclusion of validated measures of nutritional status is warranted in order to develop a typology of high risk criteria for this complex patient group. Consideration should also be given to the clinical utility of methods to assess nutritional status and skeletal muscle status in routine practice.
Funding
Chief Investigator (M.F.) was supported by a Sydney Research PhD Scholarship through the Sydney Local Health District.
Conflicts of interest
The authors declare no conflict of interest.
Author contributions
The author's contribution was as follows: Conceptualization, M.F., K.W., and J.B.; methodology, M.F., K.W., C.B., and J.B.; software, M.F.; validation, M.F. and J.B.; formal analysis, M.F. and C.B.; investigation, M.F.; resources, M.F.; data curation, M.F.; writing—original draft preparation, M.F.; writing—review and editing, M.F., K.W., C.B., and J.B.; visualization, M.F.; supervision, K.W., C.B., and J.B.; project administration, M.F.; funding acquisition, M.F., K.W., and J.B. All authors have read and agreed to the published version of the manuscript and confirm compliance with the Ethical guidelines for authorship and publishing in this journal. 36
Acknowledgements
We thank Professor Michael Fulham, Ms Patricia Knebl, and staff in the Department of Molecular Imaging, Royal Prince Alfred Hospital, Sydney New South Wales, Australia. We also thank Ms Dan Luo and Ms Michelle Lai, Cancer Nursing Research Unit at the University of Sydney for research officer support.
Findlay M., White K., Brown C., and Bauer J. D. (2021) Nutritional status and skeletal muscle status in patients with head and neck cancer: impact on outcomes, Journal of Cachexia, Sarcopenia and Muscle, 12, 2187–2198, 10.1002/jcsm.12829
References
- 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394–424. [DOI] [PubMed] [Google Scholar]
- 2. Findlay M, White K, Lai M, Luo D, Bauer JD. The association between computed tomography–defined sarcopenia and outcomes in adult patients undergoing radiotherapy of curative intent for head and neck cancer: a systematic review. J Acad Nutr Diet 2020;120:1330–1347. [DOI] [PubMed] [Google Scholar]
- 3. van Bokhorst‐de van der Schueren MA, van Leeuwen PA, Kuik DJ, Klop WM, Sauerwein HP, Snow GB, et al. The impact of nutritional status on the prognoses of patients with advanced head and neck cancer. Cancer 1999;86:519–527. [PubMed] [Google Scholar]
- 4. Marshall KM, Loeliger J, Nolte L, Kelaart A, Kiss NK. Prevalence of malnutrition and impact on clinical outcomes in cancer services: a comparison of two time points. Clin Nutr 2019;38:644–651. [DOI] [PubMed] [Google Scholar]
- 5. Baracos VE, Martin L, Korc M, Guttridge DC, Fearon KCH. Cancer‐associated cachexia. Nat Rev Dis Primers 2018;4:17105. [DOI] [PubMed] [Google Scholar]
- 6. Fearon K, Strasser F, Anker SD, Bosaeus I, Bruera E, Fainsinger RL, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol 2011;12:489–495. [DOI] [PubMed] [Google Scholar]
- 7. Fearon K, Arends J, Baracos V. Understanding the mechanisms and treatment options in cancer cachexia. Nat Rev Clin Oncol 2013;10:90–99. [DOI] [PubMed] [Google Scholar]
- 8. Cederholm T, Jensen GL, Correia M, Gonzalez MC, Fukushima R, Higashiguchi T, et al. GLIM criteria for the diagnosis of malnutrition—a consensus report from the global clinical nutrition community. Clin Nutr 2019;38:1–9. [DOI] [PubMed] [Google Scholar]
- 9. Ryan AM, Power DG, Daly L, Cushen SJ, Ní Bhuachalla Ē, Prado CM. Cancer‐associated malnutrition, cachexia and sarcopenia: the skeleton in the hospital closet 40 years later. P Nutr Soc 2016;75:199–211. [DOI] [PubMed] [Google Scholar]
- 10. Martin L, Gioulbasanis I, Senesse P, Baracos VE. Cancer‐associated malnutrition and CT‐defined sarcopenia and myosteatosis are endemic in overweight and obese patients. J Parenter Enteral Nutr 2020;44:227–238. [DOI] [PubMed] [Google Scholar]
- 11. Findlay M, White K, Stapleton N, Bauer J. Is sarcopenia a predictor of prognosis for patients undergoing radiotherapy for head and neck cancer? A meta‐analysis. Clin Nutr 2021;40:1711–1718. [DOI] [PubMed] [Google Scholar]
- 12. Hua X, Liu S, Liao J‐F, Wen W, Long Z‐Q, Lu Z‐J, et al. When the loss costs too much: a systematic review and meta‐analysis of sarcopenia in head and neck cancer. Front Oncol 2020;9:1561–1561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Wong A, Zhu D, Kraus D, Tham T. Radiologically defined sarcopenia affects survival in head and neck cancer: a meta‐analysis. Laryngoscope 2021;131:333–341. [DOI] [PubMed] [Google Scholar]
- 14. Shen W, Punyanitya M, Wang Z, Gallagher D, St‐Onge MP, Albu J, et al. Total body skeletal muscle and adipose tissue volumes: estimation from a single abdominal cross‐sectional image. J Appl Physiol 2004;97:2333–2338. [DOI] [PubMed] [Google Scholar]
- 15. Prado CM, Heymsfield SB. Lean tissue imaging: a new era for nutritional assessment and intervention. J Parenter Enteral Nutr 2014;38:940–953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. MacDonald AJ, Greig CA, Baracos V. The advantages and limitations of cross‐sectional body composition analysis. Curr Opin Support Palliat Care 2011;5:342–349. [DOI] [PubMed] [Google Scholar]
- 17. Aubrey J, Esfandiari N, Baracos VE, Buteau FA, Frenette J, Putman CT, et al. Measurement of skeletal muscle radiation attenuation and basis of its biological variation. Acta Physiologica 2014;210:489–497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Findlay M, Brown C, De Abreu Lourenço R, White K, Bauer J. Sarcopenia and myosteatosis in patients undergoing curative radiotherapy for head and neck cancer: Impact on survival, treatment completion, hospital admission and cost. J Hum Nutr Diet 2020;33:811–821. [DOI] [PubMed] [Google Scholar]
- 19. Martin L, Hopkins J, Malietzis G, Jenkins JT, Sawyer MB, Brisebois R, et al. Assessment of computed tomography (CT)‐defined muscle and adipose tissue features in relation to short‐term outcomes after elective surgery for colorectal cancer: a multicenter approach. Ann Surg Oncol 2018;25:2669–2680. [DOI] [PubMed] [Google Scholar]
- 20. Kubrak C, Martin L, Gramlich L, Scrimger R, Jha N, Debenham B, et al. Prevalence and prognostic significance of malnutrition in patients with cancers of the head and neck. Clin Nutr 2020;39:901–909. [DOI] [PubMed] [Google Scholar]
- 21. Ottery FD. Patient generated‐subjective global assessment. In McCallum P, Polisena C, eds. The Clinical Guide to Oncology Nutrition. Chicago: The American Dietetic Association; 2000. p 11–23. [Google Scholar]
- 22. Findlay M, Bauer J, Brown T, Davidson W, Isenring E, Kiss N, Kurmis R, Loeliger J, Sandison A, Talwar B, Taylor K. Evidence‐based practice guidelines for the nutritional management of adult patients with head and neck cancer. Sydney: Cancer Council Australia. http://wiki.cancer.org.au/australia/COSA:Head_and_neck_cancer_nutrition_guidelines (accessed 18 March 2021). [DOI] [PubMed]
- 23. Findlay M, Rankin NM, Shaw T, White K, Boyer M, Milross C, et al. Best evidence to best practice: implementing an innovative model of nutrition care for patients with head and neck cancer improves outcomes. Nutrients 2020;12:1465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. New South Wales Government . New South Wales Registry of Births, Deaths and Marriages. Australia; 2020. [Google Scholar]
- 25. Australian Standard Classification of Cultural and Ethnic Groups (ASCCEG) . 2019. Australian Bureau of Statistics 1249.0. https://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/1249.0Main+Features12019?OpenDocument
- 26. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata‐driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009, Apr;42(2):377–381. 10.1016/j.jbi.2008.08.010. Epub 2008 Sep 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Mishra A, Bigam KD, Extermann M, Faramand R, Thomas K, Pidala JA, et al. Sarcopenia and low muscle radiodensity associate with impaired FEV1 in allogeneic haematopoietic stem cell transplant recipients. J Cachexia Sarcopenia Muscle 2020;11:1570–1579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Martin L, Birdsell L, Macdonald N, Reiman T, Clandinin MT, McCargar LJ, et al. Cancer cachexia in the age of obesity: skeletal muscle depletion is a powerful prognostic factor, independent of body mass index. J Clin Oncol 2013;31:1539–1547. [DOI] [PubMed] [Google Scholar]
- 29. R Foundation for Statistical Computing, 1.2.5042 . R: a language and environment for statistical computing; Vienna, Austria, 2017.
- 30. Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A, editors. AJCC Cancer Staging Manual, (7th ed). New York, NY: Springer; 2010. https://www.springer.com/gp/book/9780387884424 [Google Scholar]
- 31. Kiss N, Loeliger J, Findlay M, Isenring E, Baguley BJ, Boltong A, et al. Clinical Oncology Society of Australia: position statement on cancer‐related malnutrition and sarcopenia. Nutrition and Dietetics 2020;77:416–425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Stretch C, Aubin JM, Mickiewicz B, Leugner D, Al‐Manasra T, Tobola E, et al. Sarcopenia and myosteatosis are accompanied by distinct biological profiles in patients with pancreatic and periampullary adenocarcinomas. PLoS ONE 2018;13:e0196235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Murnane LC, Forsyth AK, Koukounaras J, Pilgrim CH, Shaw K, Brown WA, et al. Myosteatosis predicts higher complications and reduced overall survival following radical oesophageal and gastric cancer surgery. Eur J Surg Oncol 2021;47:2295–2303. [DOI] [PubMed] [Google Scholar]
- 34. Xiao J, Caan BJ, Cespedes Feliciano EM, Meyerhardt JA, Peng PD, Baracos VE, et al. Association of low muscle mass and low muscle radiodensity with morbidity and mortality for colon cancer surgery. JAMA Surg 2020;155:942–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Boshier PR, Heneghan R, Markar SR, Baracos VE, Low DE. Assessment of body composition and sarcopenia in patients with esophageal cancer: a systematic review and meta‐analysis. Dis Esophagus 2018;31:1–11. [DOI] [PubMed] [Google Scholar]
- 36. von Haehling S, Morley JE, Coats AJS, Anker SD. Ethical guidelines for publishing in the journal of cachexia, sarcopenia and muscle: update 2019. J Cachexia Sarcopenia Muscle 2019;10:1143–1145. [DOI] [PMC free article] [PubMed] [Google Scholar]