Abstract

Background: The purpose of the present study was to examine the association between non-exercise Cardiorespiratory Fitness (eCRF) and all-cause mortality in cancer patients.

Methods and Findings: A total of 2,404 participants from the Aerobics Center Longitudinal Study (622 women and 1,782 men) with a cancer diagnosis were followed for mortality. Non-exercise eCRF was calculated in Metabolic Equivalents (METs) with sex-specific algorithms at baseline.  Multivariable Cox regression models were used to examine the association between CRF and risk of all-cause mortality.  Hazard ratios and 95% confidence intervals were calculated as an index of strength of the association. More than 96% (2,323) of the 2,404 participants survived, while 81 (3.4%) died. In the multivariable adjusted model, each 1-MET increment was associated with a 17% decreased risk of all-cause mortality. Compared with the reference group, those in the middle CRF group had a 58% lower risk of death and those in the upper CRF group had a 78% lower risk of death than those in the lower CRF group (Ptrend=0.0002).

Conclusion: CRF estimated using a non-exercise formula (non-exercise eCRF) is inversely associated with all-cause mortality in cancer patients.

Keywords: Cardiorespiratory Fitness; Non-Exercise; Mortality; Cancer Survivors

Introduction

Cancer is one of the leading causes of death on a global scale, accounting for approximately 10 million deaths in 2020 or one in six deaths in that year.  Between 2015 and 2050, it is projected, partly due to the increasing growth and age of the world population, that the annual number of cancer cases will increase 49%, from 1,534,500 in 2015 to 2,286,300 in 2050 [1]. Additionally, nearly one-third of cancer deaths can be traced to tobacco use, high Body Mass Index (BMI), alcohol consumption, poor diet, and lack of Physical Activity (PA) [2].

These risk factors, among others, are also major determinants of Cardiorespiratory Fitness (CRF), which is a measure of the capacity of the cardiopulmonary system to effectively transport oxygen and nutrients to active skeletal muscles for adenosine triphosphate resynthesis.  It is essentially an objective estimation of a person’s physiological functioning [3]. CRF has been established as a predictor of mortality from Cardiovascular (CV) disease (CVD) and it is usually measured via an Exercise Tolerance Test (ETT) in a healthcare setting. Therefore, the American Heart Association has proposed it as a clinical vital sign due to overwhelming evidence of its application to general CV health [4].  However, research has evaluated the accuracy of non-exercise estimated CRF (eCRF) as an alternative to exercise-based and its significance as a predictor of mortality from various diseases.  This is primarily due to the cost and time burdens involved in exercise tests [5].

Recent studies have demonstrated significant relationships between non-exercise eCRF and health outcomes, including mortality and incident hypertension, stroke and depression [5-8]. However, no prior study has investigated the association between non-exercise eCRF and all-cause mortality in cancer patients. We previously demonstrated that PA, a major determinant of CRF, was not associated with cancer survival, whereas Resistance Exercise (RE) was a significant predictor [9]. A prior analysis from the same cohort, however, demonstrated that higher CRF measured by ETT was a predictor of cancer survival [10]. Since ETT is not routinely performed clinically in cancer survivors to more precisely measure CRF, the purpose of this study was to examine the association between non-exercise eCRF and all-cause mortality among a group of cancer survivors.   Existing data on the patient population, including men and women, was taken from the Aerobics Center Longitudinal Study (ACLS).  The non-exercise estimation was derived from a formula using readily available medical records, including variables such as BMI, Resting Heart Rate (RHR) and Blood Pressure (BP) [11]. The formula could potentially be used as a valid alternative for assessing the risk of all-cause mortality in cancer patients.

Methods

Study Design and Participants

We used data from the ACLS where adult men and women were examined at the Cooper Clinic in Dallas, Texas. The ACLS is a longitudinal observation study in which new and recurring patients visited the clinic for a comprehensive physical examination and for lifestyle counseling, such as exercise, nutrition and stress management, since the 1970s. The current study included cancer participants at baseline between 1987 and 2002 as well as complete data to calculate the non-exercise eCRF. They were then followed for mortality after baseline clinical visit. These individuals, at the time of testing, all reported a history of cancer on their medical history questionnaire. Patients came from all 50 states and were volunteers, sent by their physicians and employers or were self-referred. Patients provided written informed consent for enrollment in the follow-up study. The Cooper Institute Institutional Review Board annually reviewed and approved the study procedures.

Non-exercise eCRF at Baseline

Non-exercise eCRF was calculated in metabolic equivalents (METs; 3.5 mL O2·kg-1·min-1) with the following two sex-specific algorithms [11]:

eCRF in men (METs) =21.2870 + (0.1654*age) – (0.0023*age²) – (0.2318*BMI) – (0.0337*waist circumference or WC) – (0.0390*RHR) + (0.6351*physically active) -(0.4263*current smoker); and eCRF in women (METs) =14.7873 + (0.1159*age) – (0.0017*age²) – (0.1534*BMI) – (0.0085*WC) – (0.0364*RHR) + (0.5987*physically active) -(0.2994*current smoker).

METs was implemented as a continuous variable and higher METs equates to better CRF. Participants were further classified into lower, middle and upper groups on the basis of age- (18-30, 40-49, 50-59 or ≥60 years) and sex-specific thirds of the estimated METs distribution. Table 1 shows the thresholds that define the 3 CRF categories.

Age, BMI, WC and RHR were entered as continuous variables, while being physically active and a current smoker were considered dichotomous variables, with a value of 1 when present and 0 when absent. BMI was calculated from measured height and weight according to standard procedures during patients’ visits. WC was measured in line with the umbilicus. Resting heart rate was determined with the participants recumbent after 5-minute rest prior to the test and was retrieved from the resting electrocardiogram. A 5-level index was used to determine PA, based on a prior validated questionnaire [12]. Walking or jogging >10 miles per week was considered physically active and was equal to index levels 3-4. Walking or jogging <10 miles per week, participating in some sport without walking or jogging or not walking or jogging at all was considered inactive and equal to index levels 0-2. Participants self-reported their smoking status as never smoked, former smoker or current smoker.

Potential Confounders

Potential confounders of the study included age, sex, alcohol consumption, hypertension, Diabetes Mellitus (DM) and hypercholesterolemia. This information was obtained during baseline examinations following the Cooper Clinic’s standardized manual of operation. Following a minimum of 12 hours fasting overnight, a physical examination and preventative health evaluation were performed. Measures of BP, a chemical analysis of blood, personal medical history and lifestyle habits were included in the examination. Hypertension (Systolic BP/Diastolic BP≥130/80 mmHg or physician diagnosis), DM (fasting glucose ≥7 mmol/L, insulin use or physician diagnosis), hypercholesterolemia (cholesterol ≥ 6.2 mmol or physician diagnosis), depression and parental history of cancer were determined from medical history.

All-Cause Mortality Determination

Vital status was ascertained using the National Death Index and death certificates from states in which participant deaths occurred. More than 95% of mortality follow-up is complete by these methods. The follow-up interval was computed from the date of a participant’s baseline examination until the date of death for decedents or until December 31, 2003, for survivors.

Statistical Analysis

Baseline characteristics of the population of patients in the study were summarized by survival status. Groups were compared using t tests and X² tests. The association between non-exercise eCRF and risk of all-cause mortality was examined using Cox regression models. Hazard Ratios (HRs) and 95% Confidence Intervals (CIs) were implemented as an index of the strength of association. Two separate models were tested: an age- and gender-adjusted model and a multivariable-adjusted model. Covariates included in the multivariable-adjusted model included: age (years), gender, year of baseline examination, heavy alcohol intake (>14 drinks/week in men or >7 drinks/week in women), presence of hypertension, DM or hypercholesterolemia, depression and parental history of cancer. We assessed linear trends in the association of non-exercise eCRF with the risk of all-cause mortality. We also examined non-exercise eCRF as a continuous variable so that each HR represents the risk associated with a 1-MET increase of eCRF. The proportional hazards assumptions were met for eCRF by comparing the log-log survival plots. Analyses were conducted in 2024 using SAS, version 9.4 (SAS, Cary, NC) with alpha set at P < 0.05.