Body Adiposity, Not BMI, Key to CV Risk in Colorectal Cancer

Liam Davenport

May 21, 2019

Patients with colorectal cancer (CRC) should have their body composition assessed on computed tomography (CT) imaging at diagnosis to assess their risk of cardiovascular disease (CVD) rather than relying on body mass index (BMI), say US investigators, as it is a better indicator of outcomes.

Justin C. Brown, PhD, director of the Cancer Metabolism Program at Pennington Biomedical Research Center, Baton Rouge, Louisiana, and colleagues looked at the records of over 2800 CRC survivors treated in the Kaiser Permanente health system.

As expected, these patients had a high rate of cardiovascular events, at a 10-year cumulative incidence of the composite major adverse cardiovascular events (MACE) of cardiovascular mortality, nonfatal myocardial infarction, and nonfatal stroke of almost 20%.

However, body mass index (BMI) was unrelated to the occurrence of MACE.

Instead, the researchers found that visceral adiposity obtained from CT imaging taken at diagnosis was significantly associated with cardiovascular outcomes.

Patients with the highest levels of abdominal fat were more than 50% more likely to experience MACE over almost 7 years of follow-up.

In addition, muscle radiodensity, which reflects the amount of lipid stored in skeletal muscle, was inversely associated with MACE, such that patients with the highest muscle density had a 33% reduced risk vs those with the lowest.

The study was published online May 16 in the journal JAMA Oncology.

Brown told Medscape Medical News that BMI "is actually quite a crude measure" for assessing CVD risk, as "it only accounts for how tall you are and how much you weigh."

He added: "The challenge with BMI is it doesn't differentiate muscle tissue from fat."

For CRC patients, who are already undergoing CT imaging at diagnosis, Brown said that "one of the rather exciting developments in this area is there are now automated techniques that can very precisely quantify the amount of muscle and fat from these CT images."

The idea is that, eventually, when patients undergo CT imaging to potentially identify a cancer, "the radiology workflow in their report will also say: here are the estimates of the visceral fat, here are the estimates of the muscle."

Brown noted: "As this study has shown, these are important metrics to report that could influence clinical decision making, because of their prognostic importance."

"We measure blood pressure, we take the pulse; my hope is that in the near future we'll be measuring these body composition metrics just like we do these other indicators of health," he added.

Brown said that "a lot of the focus to date has been on the risk of cancer occurrence, and that certainly is of great importance because that is the number one cause of morbidity and mortality in this population, but number two is cardiovascular disease."

"This study provides further evidence that patients, after their diagnosis, should engage in physical activity, should eat a healthy diet, and maintain a healthy body composition, in order to try to minimize their risk of experiencing a cardiovascular event," he said.

In an accompanying editorial, Michael N. Passarelli, PhD, Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, agrees with the authors.

While BMI "will still remain useful in the general population to assess obesity," particularly in primary care, he agreed with Brown that the ready availability of CT imaging for cancer patients means it "could be easily repurposed" to determine body composition.

The results may also help to explain the so-called obesity paradox, in which individuals — including those with CRC — who are overweight or in the early stages of obesity are less likely to die than normal weight individuals.

"I think the takeaway here, at least from my point of view, is that the notion of the obesity paradox for mortality from colorectal cancer and separately from cardiovascular disease seems to be largely an artifact of the inadequacies of measuring obesity using height and weight," Passarelli told Medscape Medical News.

"Dr [Bette] Caan, who's one of the authors of this paper, has come out and said that, really, the obesity paradox is itself a BMI paradox, and I think that's a very apt description," he said.

Passarelli explained: "For two people with the exact same height and weight, it's obvious that there's substantial heterogeneity in their muscle mass and where they carry their weight, and we know that this variation can inform risk of developing or risk of fatal colorectal cancer and cardiovascular disease."

Study Details

Survival among CRC patients has increased by 33% over the past 4 decades, but these patients are now more susceptible to competing causes of morbidity and mortality.

Indeed, recent studies have shown that CRC patients have a two- to fourfold increased risk of developing CVD than the general population.

This has led to recommendations that CRC patients be counseled on the association between BMI-defined obesity and CVD risk, a notion that is based entirely on expert opinion.

Against a background of uncertainty over the utility of BMI in determining cardiovascular risk in the general population, and the availability of algorithms to assess body composition on CT scans, the researchers looked at whether composition would be a better indicator of cardiovascular event risk in CRC.

They conducted a retrospective analysis of patients from the Kaiser Permanente Northern California cancer registry who were diagnosed with stage I to III CRC between 2006 and 2011.

All patients had undergone abdominal or pelvic CT scanning at diagnosis and surgical resection of their CRC, and had valid body mass measures. Patients with a history of myocardial infarction or stroke prior to CRC diagnosis were excluded.

Baseline information on age, sex, self-reported race and ethnicity, and CVD risk factors such as smoking history, hypertension, hyperlipidemia, and type 2 diabetes were collated.

The researchers then gathered information from the California State death registry and the National Death Index, and determined the time to the first occurrence of MACE.

A total of 2839 patients (51.2% women; average age 61.9 years) were included. The majority of patients were former or current smokers (52%) and had hypertension (55%), while a substantial proportion had hyperlipidemia (49%) and type 2 diabetes (20%).

Over a median follow-up of 6.8 years, MACE occurred in 366 (12.8%) patients, at a cumulative incidence of 3.4%, 5.9%, and 19.1% at 1, 3, and 10 years, respectively.

CT analysis showed that BMI was positively correlated with visceral adiposity, subcutaneous adiposity, and muscle mass, and negatively correlated with muscle radiodensity.

However, BMI was not associated with the risk of MACE, at a nonsignificant hazard ratio for 35 kg/m2 vs 18.5–24.9 kg/m2 of 1.30 (P = .50).

Visceral adiposity was, in contrast, was significantly associated with MACE, at a multivariate adjusted hazard ratio for the lowest vs highest quintile of 1.54 (P = .04).

Muscle radiodensity was also significantly inversely associated with MACE, at a hazard ratio for the highest vs lowest quintile of 0.67 (P = .02).

Subcutaneous adiposity and muscle mass were not associated with the risk of MACE, and the associations were not affected by accounting for sex.

Brown, the study's lead author, said that the association between visceral adiposity and low muscle radiodensity and cardiovascular outcomes is likely due to the adiposity being in the wrong place.

He explained that visceral fat does not "belong there in the first place, so it initiates this cascade of inflammation and insulin resistance, and basically the body begins to go haywire, where it doesn't know what to do anymore."

Passarelli added that both obesity in general and visceral adiposity are also risk factors for developing CRC, which compounds the issue.

He nevertheless suspects that "these results are generalizable to cardiovascular outcomes in patients that have other cancers, or perhaps even patients without cancer."

In the editorial, Passarelli calls for large-scale epidemiologic studies to look at individual, as opposed to composite, cardiovascular endpoints to tease apart the associations with body composition.

It can be done, he said, as "all of this imaging data is routinely collected, and so it would be pretty easy to scale up to even larger studies."

"Every individual that has cancer is getting imaging routinely to assess their stage, so it's there, it's part of the medical record, so I can foresee much larger studies that to distinguish more specific endpoints," he elaborated.

Research reported in this study was supported by the National Cancer Institute of the National Institutes of Health. Brown reports grants from the National Cancer Institute (paid to his institution, Pennington Biomedical Research Center). Study coauthor Carla M. Prado reports personal fees from Abbott Nutrition outside the submitted work. Study coauthor Elizabeth M. Cespedes Feliciano reports grants from National Cancer Institute during the conduct of the study. Passarelli has disclosed no relevant financial relationships.

JAMA Oncol. Published online May 16, 2019. Abstract, Editorial

For more from Medscape Oncology, follow us on Twitter:  @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.