The mortality burden of cachexia or weight loss in patients with colorectal or pancreatic cancer: A systematic literature review

Abstract

Cancer-associated cachexia is a multifactorial wasting disorder characterized by anorexia, unintentional weight loss (skeletal muscle mass with or without loss of fat mass), progressive functional impairment, and poor prognosis. This systematic literature review (SLR) examined the relationship between cachexia and survival in patients with colorectal or pancreatic cancer in recent literature. The SLR was conducted following PRISMA guidelines. Embase^® and PubMed were searched to identify articles published in English between 1 January 2016 and 10 October 2021 reporting survival in adults with cancer and cachexia or at risk of cachexia, defined by international consensus (IC) diagnostic criteria or a broader definition of any weight loss. Included publications were studies in ≥100 patients with colorectal or pancreatic cancer. Thirteen publications in patients with colorectal cancer and 13 with pancreatic cancer met eligibility criteria. Included studies were observational and primarily from Europe and the United States. Eleven studies (42%) reported cachexia using IC criteria and 15 (58%) reported any weight loss. An association between survival and cachexia or weight loss was assessed across studies using multivariate (n = 23) or univariate (n = 3) analyses and within each study across multiple weight loss categories. Cachexia/weight loss was associated with a statistically significantly poorer survival in at least one weight loss category in 16 of 23 studies that used multivariate analyses and in 1 of 3 studies (33%) that used univariate analyses. Of the 17 studies demonstrating a significant association, 9 were in patients with colorectal cancer and 8 were in patients with pancreatic cancer. Cachexia or weight loss was associated with significantly poorer survival in patients with colorectal or pancreatic cancer in nearly two-thirds of the studies. The classification of weight loss varied across and within studies (multiple categories were evaluated) and may have contributed to variability. Nonetheless, awareness of cachexia and routine assessment of weight change in clinical practice in patients with colorectal or pancreatic cancer could help inform prognosis and influence early disease management strategies.

Keywords: Cachexia; Colorectal cancer; Muscle wasting; Pancreatic cancer; Systematic literature review; Weight loss.

Figure 1

PRISMA flow diagram for identification of relevant studies for the colorectal or pancreatic cancer SLR. ^aA total of 25 individual publications were identified, but 1 publication (Gannavarapu et al., 2018 ²⁰ ) reported weight‐loss and survival data for colorectal and pancreatic cancer populations separately and was counted twice. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta‐Analyses; SLR, systematic literature review; WL, weight loss. Reproduced (with modifications) from Bonomi PD et al. Mortality burden of pre‐treatment weight loss in patients with non‐small‐cell lung cancer: A systematic literature review and meta‐analysis. Journal of Cachexia, Sarcopenia and Muscle, 2024 Apr 22 [19] licensed under CC BY 4.0.

Figure 2

Prevalence of cachexia (IC criteria or weight loss ≥5%) in patients with (A) colorectal cancer and (B) pancreatic cancer. ^aNaumann P, et al. Cancers (Basel) 2019;11:1655. ^bNaumann P, et al. Cancers (Basel) 2019;11:709.

Figure 3

Association between cachexia or weight loss and overall survival from multivariate analyses in colorectal cancer studies. Of the 12 multivariate analyses in studies on colorectal cancer, 11 studies are shown: The remaining study (Gannavarapu et al., 2018 ²⁰ ) did not report an HR, but identified a significant association between WL ≥ 5% and worse survival. Overall, nine studies using multivariate analyses showed that for at least one category assessed, cachexia or WL was associated with a statistically significant poorer survival in patients with colorectal cancer. The 1 univariate analysis (Karabulut et al., 2021 ³⁹ ) showed no difference in survival for WL > 10% versus <10%. Studies have been grouped by cachexia definition, WL category, and WL grade. ^aHR was inversed to present severe WL versus no WL (reference category). ^bGrade 0: BMI ≥ 25 and WL < 2.5%; Grade 1: BMI 20 to <25 and WL < 2.5% or BMI ≥ 28 and WL 2.5% to <6%; Grade 2: BMI 20 to <28 and WL 2.5% to <6% or BMI ≥ 28 and WL 6% to <11%; Grade 3: BMI < 20 and WL < 6% or BMI 20 to <28 and WL 6% to <11%; Grade 4: BMI < 20 and WL ≥ 6% or BMI 20 to <22 and WL ≥ 11% or BMI 22 to <28 and WL ≥ 15%. Data are shown on a log₂ scale. BMI, body mass index; CI, confidence interval; HR, hazard ratio; mo, months; NR, not reported; NS, not significant; Tx, treatment; wks, weeks; WL, weight loss.

Figure 4

Association between cachexia or weight loss and overall survival from multivariate analyses in pancreatic cancer studies. Of the 11 multivariate analyses in studies on pancreatic cancer, nine studies are shown. The remaining two studies (Arthur et al., 2016 and Gannavarapu et al., 2018 ²⁰ ) did not report an HR, and both had non‐significant results. Overall, seven studies using multivariate analyses showed that for at least one category assessed, cachexia or WL was associated with a statistically significantly poorer survival for patients with pancreatic cancer. Of the two studies (Latenstein et al., 2020 and Domínguez‐Muñoz et al., 2018 ⁴⁵ ) that used univariate analyses, one (Latenstein et al., 2020 ¹⁸ ) demonstrated a significant association. Studies have been grouped by cachexia definition and WL category.^aNaumann P, et al. Cancers (Basel) 2019;11:1655. ^bNaumann P, et al. Cancers (Basel) 2019;11:709. Data are shown on a log₂ scale. CI, confidence interval; HR, hazard ratio; CRT, chemoradiotherapy; CWL, continuous weight loss; Tx, treatment; wk, week; WL, weight loss.