Abstract
Aims
The prognostic importance of post-diagnosis assessment of cardiorespiratory fitness (CRF) in cancer patients is not well established. We sought to examine the association between CRF and mortality in cancer patients.
Methods and results
This was a single-centre cohort analysis of 1632 patients (58% male; 64 ± 12 years) with adult-onset cancer who were clinically referred for exercise treadmill testing a median of 7 [interquartile range (IQR): 3–12] years after primary diagnosis. Cardiorespiratory fitness was defined as peak metabolic equivalents (METs) achieved during standard Bruce protocol and categorized by tertiles. The association between CRF and all-cause and cause-specific mortality was assessed using multivariable Cox proportional hazard models adjusting for important covariates. Median follow-up was 4.6 (IQR: 2.6–7.0) years; a total of 411 deaths (229, 50, and 132 all-cause, cardiovascular (CV), and cancer related, respectively) occurred during this period. Compared with low CRF (range: 1.9–7.6 METs), the adjusted hazard ratio (HR) for all-cause mortality was 0.38 [95% confidence interval (CI): 0.28–0.52] for intermediate CRF (range: 7.7–10.6 METs) and 0.17 (95% CI: 0.11–0.27) for high CRF (range: 10.7–22.0 METs). The corresponding HRs were 0.40 (95% CI: 0.19–0.86) and 0.41 (95% CI: 0.16–1.05) for CV mortality and 0.40 (95% CI: 0.26–0.60) and 0.16 (95% CI: 0.09–0.28) for cancer mortality, respectively. The adjusted risk of all-cause, CV, and cancer mortality decreased by 26%, 14%, and 25%, respectively with each one MET increment in CRF.
Conclusion
Cardiorespiratory fitness is a strong, independent predictor of all-cause, CV, and cancer mortality, even after adjustment for important clinical covariates in patients with certain cancers.
Keywords: Cardiorespiratory fitness, Exercise, Cancer, Cardio-oncology, Mortality, Cancer mortality, Cause-specific mortality, Cardiovascular mortality, Cardiac rehab
Introduction
A plethora of demographic, medical, biochemical, and genomic markers are used for risk stratification and prognostication in daily oncology practice. Of these, performance status, an assessment of patient physical functioning and capability of self-care, is consistently a strong independent predictor in numerous cancer settings.1–7 Accordingly, performance status is frequently considered in decisions regarding initial eligibility and tolerability of selected anticancer regimens. However, performance status scoring systems are subjective, with poor inter-rater reliability8 stimulating investigation of alternative tools (e.g. comprehensive geriatric assessment,9 frailty measures,10 hematopoietic cell transplantation-comorbidity index11) postulated to provide a more objective assessment of physical functioning. Nevertheless, all current tools are still largely subjective in nature, failing to fully characterize global physical functioning under conditions of stress. Consequently, sensitivity is poor, particularly in patients presenting with ‘good’ performance status.12,13
Cardiorespiratory fitness (CRF), as measured by an exercise tolerance test (ETT), reflects the integrated capacity of the cardiopulmonary system to deliver adequate oxygen and substrate to active skeletal muscles for adenosine triphosphate resynthesis. Cardiorespiratory fitness, therefore, provides a robust, objective evaluation of global physical functioning.14 Cardiorespiratory fitness is an established independent predictor of death from cardiovascular (CV) disease and death from any cause in a broad range of adult populations.15–18 Accordingly, formalized ETT is an integral tool in numerous clinical settings.19
In stark contrast, the measurement of CRF is not a routine aspect of clinical care in patients with cancer. From a research perspective, measurement of CRF via ETT is utilized to evaluate the efficacy of exercise interventions in cancer patients; however, whether CRF has prognostic importance has received minimal attention.14 We determined the association between post-diagnosis CRF and all-cause and cause-specific mortality in a diverse cohort of cancer patients referred clinically for ETT. We hypothesized that high CRF reduces the risk of death from all causes after adjustment for important clinical covariates.
Methods
Study population and design
This was a single centre, retrospective cohort study. Using the Partners Healthcare Research Patient Data Registry, 28 959 consecutive patients completing an ETT for clinical indications between November 2001 and August 2015 were identified. Using ICD-9 codes, case records were further manually screened for a diagnosis of prostate, breast, Hodgkin lymphoma, or head and neck cancer. These patients were previously identified20,21 and combined for the present analysis. A total of 2827 unique patients were identified. Patients in whom a cancer diagnosis could not be confirmed, or diagnosis occurred after index ETT, had incomplete cancer diagnosis and treatment information, or the ETT did not follow standardized procedures were excluded, resulting in a final analytic cohort of 1632 patients (Figure 1). This study was approved by the Partners Healthcare Institutional Review Board.
Figure 1.
Cohort composition diagram of eligible and enrolled cancer patients.
Exercise tolerance test protocol and cardiorespiratory fitness assessment
A formalized ETT was conducted per standard Bruce Protocol procedures.22 Heart rate and blood pressure (BP) were recorded at rest, after each 3-min stage of exercise, at maximum exercise, and at 1, 3, and 5 min in recovery. Exercise continued until ≥1 of the following endpoints: exhaustion, symptom limitation, ≥85% of age-predicted maximal heart rate (220-age), ≥10 mmHg drop in systolic BP from baseline, or significant electrocardiogram (EKG) abnormalities were detected. Exercise tolerance tests were performed, analysed, and reported per international standards23,24 using a computerized database. Resting left ventricular ejection fraction (LVEF) was recorded for ETTs performed with echocardiography or nuclear myocardial perfusion imaging. Cardiorespiratory fitness was expressed in units of metabolic equivalents (METs), calculated from peak treadmill speed and grade. One MET is defined as the energy expended while sitting quietly, which is equivalent to an oxygen consumption of ∼3.5 mL/kg/min.25 Patients were categorized as low, intermediate, and high CRF based on tertiles of METs achieved during ETT.
Ascertainment of deaths
The primary endpoint was all-cause mortality. Secondary endpoints were CV and cancer mortality. All-cause and cause-specific mortality through December 2015 were determined using the National Death Index.
Clinical history
Demographics, indications for exercise testing, CV history, and medications were abstracted at the time of ETT by a structured patient interview and electronic medical record review. Ischaemic heart disease was defined as prior myocardial infarction, coronary revascularization, or documented angiographic coronary artery disease. Congestive heart failure included history of heart failure, cardiomyopathy, LVEF <40%, or loop diuretic use. Hyperlipidaemia was defined as history of hyperlipidaemia or statin use. Diabetes was defined as history of diabetes or use of insulin or oral hypoglycaemic agents. A positive smoking history included ongoing or prior tobacco use. Body mass index (BMI) was calculated using patients’ height and weight. The Morise score was calculated for each patient to assess the pre-test probability of a positive stress test.26 Cancer medical history was also abstracted.
Statistical analysis
Categorical variables are presented as percentages and compared using the χ2 test. Continuous, normally distributed variables are presented as mean ± standard deviation (SD) and compared using analysis of variance. Continuous, non-normal data are presented as median with interquartile range (IQR) and compared using the Wilcoxon rank-sum test. Cumulative incidence of all-cause and cause-specific death at 5 and 10 years from ETT was stratified by tertiles of CRF and compared using permutation tests. When cause-specific cumulative incidence of death was estimated, all other causes of death were treated as competing risk events. The at-risk time was censored at the time of National Death Index search. Cox proportional-hazards regression analysis was used to examine the association between CRF and mortality rates. All-cause, CV, and cancer mortality were compared between tertiles of CRF using Kaplan–Meier curves. Unadjusted and adjusted HRs with 95% confidence intervals (CIs) are presented. Models were adjusted for age, gender, race, BMI, use of atrioventricular nodal blocking agents, Morise score, result of ETT, and interval from cancer diagnosis to ETT. A two-sided P-value of <0.05 was considered significant. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).
Results
Study cohort
Median time from primary cancer diagnosis to ETT was 7 (IQR: 3–12) years. For the overall cohort, mean CRF was 9.6 ± 3.2 METs. Cardiorespiratory fitness was divided into tertiles defined as low (range: 1.9–7.6 METs, mean 6.1 ± 1.1 METs), intermediate (range: 7.7–10.6 METs, mean 9.4 ± 0.8 METs), and high (range: 10.7–22.0 METs, mean 13.2 ± 2.1 METs). Baseline characteristics of the study cohort by CRF tertile are listed in Table 1. Compared to the low CRF tertile, high CRF patients were younger at diagnosis and ETT, more likely to be Caucasian and had a more favourable CV risk profile (Table 1).
Table 1.
Demographic and treatment characteristics of patients according to cardiorespiratory fitness tertiles
| METs |
|||||
|---|---|---|---|---|---|
| Characteristics | Overall | Low CRF | Intermediate CRF | High CRF | P-valuea |
| Number of participants (%) | 1632 (100) | 547 (33.5) | 548 (33.6) | 537 (32.9) | |
| Age at ETT (years) | <0.001 | ||||
| Mean (SD) | 64 (12) | 69 (10) | 64 (11) | 57 (11) | |
| Median (IQR) | 65 (56–72) | 70 (63–76) | 65 (58–72) | 59 (49–66) | |
| Age at diagnosis (years) | <0.001 | ||||
| Mean (SD) | 55 (15) | 61 (13) | 56 (14) | 47 (16) | |
| Median (IQR) | 58 (47–66) | 63 (54–69) | 58 (49–65) | 51 (35–60) | |
| Time from diagnosis to ETT (years) | <0.001 | ||||
| Mean (SD) | 9 (8) | 8 (8) | 9 (8) | 10 (8) | |
| Median (IQR) | 7 (3–12) | 6 (3–11) | 6 (3–12) | 8 (4–14) | |
| Males, n (%) | 954 (58) | 321 (59) | 321 (59) | 312 (58) | 0.85 |
| Race/ethnicity, n (%) | <0.001 | ||||
| Non-Hispanic White | 1340 (82) | 419 (77) | 443 (81) | 478 (89) | |
| Other | 292 (18) | 128 (23) | 105 (19) | 59 (11) | |
| Cancer type, n (%) | <0.001 | ||||
| Head and neck cancer | 208 (13) | 73 (13) | 69 (13) | 66 (12) | |
| Hodgkin lymphoma | 253 (16) | 35 (6) | 63 (11) | 155 (29) | |
| Breast cancer | 436 (27) | 169 (31) | 148 (27) | 119 (22) | |
| Prostate cancer | 735 (45) | 270 (49) | 268 (49) | 197 (37) | |
| Cancer therapy, n (%) | |||||
| Any chemotherapy | 0.23 | ||||
| Yes | 1131 (69) | 382 (70) | 394 (72) | 355 (66) | |
| No | 478 (29) | 154 (28) | 148 (27) | 176 (33) | |
| Unknown | 23 (1) | 11 (2) | 6 (1) | 6 (1) | |
| Anthracyclines | <0.001 | ||||
| Yes | 1335 (82) | 464 (85) | 465 (85) | 406 (76) | |
| No | 267 (16) | 69 (13) | 75 (14) | 123 (23) | |
| Unknown | 30 (2) | 14 (3) | 8 (1) | 8 (1) | |
| Non-anthracyclines | 0.004 | ||||
| Yes | 1408 (86) | 456 (83) | 472 (86) | 480 (89) | |
| No | 201 (12) | 80 (15) | 70 (13) | 51 (9) | |
| Unknown | 23 (1) | 11 (2) | 6 (1) | 6 (1) | |
| Supradiaphragmatic RT | 0.006 | ||||
| Yes | 907 (56) | 314 (57) | 330 (60) | 263 (49) | |
| No | 703 (43) | 225 (41) | 212 (39) | 266 (50) | |
| Unknown | 22 (1) | 8 (1) | 6 (1) | 8 (1) | |
| Chemotherapy and supradiaphragmatic RT | 0.15 | ||||
| Yes | 1226 (75) | 415 (76) | 426 (78) | 385 (72) | |
| No | 386 (24) | 122 (22) | 115 (21) | 149 (28) | |
| Unknown | 20 (1) | 10 (2) | 7 (1) | 3 (1) | |
| Hormonal therapy | <0.001 | ||||
| Yes | 1171 (72) | 351 (64) | 389 (71) | 431 (80) | |
| No | 446 (27) | 191 (35) | 152 (28) | 103 (19) | |
| Unknown | 15 (1) | 5 (1) | 7 (1) | 3 (1) | |
| BMI (kg/m2) | <0.001 | ||||
| Mean (SD) | 27.2 (4.9) | 28.5 (5.3) | 27.4 (4.8) | 25.5 (4.1) | |
| Median | 26.6 | 28.0 | 26.7 | 25.3 | |
| IQR | (23.9–29.7) | (24.8–31.6) | (24.1–29.8) | (22.8–27.8) | |
| CV risk factors | |||||
| Smoking, n (%) | 99 (6) | 47 (9) | 35 (6) | 17 (3) | <0.001 |
| Diabetes mellitus, n (%) | 249 (15) | 130 (24) | 96 (18) | 23 (4) | <0.001 |
| Hypertension, n (%) | 921 (56) | 390 (71) | 322 (59) | 209 (39) | <0.001 |
| Dyslipidaemia, n (%) | 992 (61) | 377 (69) | 350 (64) | 265 (49) | <0.001 |
| Ischaemic heart disease, n (%) | 345 (21) | 145 (27) | 123 (23) | 77 (14) | <0.001 |
| CHF, n (%) | 131 (8) | 83 (15) | 33 (6) | 15 (3) | <0.001 |
| LVEF (%)b | 58 (8) | 57 (10) | 59 (7) | 59 (6) | 0.005 |
| Morise score | 11.9 (3.8) | 13.5 (2.9) | 12.2 (3.4) | 9.9 (4.1) | <0.001 |
| Cardiac meds, n (%) | |||||
| Beta-blocker | 634 (39) | 292 (53) | 220 (40) | 122 (23) | <0.001 |
| Ca++ channel blocker | 230 (14) | 120 (22) | 68 (12) | 42 (8) | <0.001 |
| AV nodal blocker | 749 (46) | 337 (62) | 264 (48) | 148 (28) | <0.001 |
| ACEi | 456 (28) | 193 (35) | 154 (28) | 109 (20) | <0.001 |
| Aspirin | 718 (44) | 271 (50) | 255 (47) | 192 (36) | <0.001 |
| Statin | 800 (49) | 305 (56) | 283 (52) | 212 (39) | <0.001 |
| Exercise time (min) | 8.0 (3.1) | 4.7 (1.3) | 8.0 (0.9) | 11.4 (1.8) | <0.001 |
| METs completed | 9.6 (3.2) | 6.1 (1.1) | 9.4 (0.8) | 13.2 (2.1) | <0.001 |
| Result of ETT, n (%) | <0.001 | ||||
| Negative | 1278 (78) | 404 (74) | 437 (80) | 437 (81) | |
| Positive | 216 (13) | 59 (11) | 81 (15) | 76(14) | |
| Inconclusive | 138 (8) | 84 (15) | 30 (5) | 24 (4) | |
| ETT parameters | |||||
| Resting heart rate (b.p.m.) | 70 (13) | 71 (13) | 70 (13) | 69 (13) | 0.002 |
| Resting SBP (mmHg) | 129 (18) | 134 (19) | 130 (17) | 125 (16) | <0.001 |
| Resting DBP (mmHg) | 76 (10) | 75 (11) | 76 (9) | 75 (9) | 0.78 |
| Peak heart rate (b.p.m.) | 142 (25) | 128 (23) | 141 (20) | 158 (20) | <0.001 |
| Peak SBP (mmHg) | 164 (25) | 160 (26) | 166 (25) | 166 (22) | <0.001 |
| Peak DBP (mmHg) | 74 (10) | 75 (10) | 74 (10) | 73 (10) | 0.002 |
ACEi, angiotensin converting enzyme inhibitor; AV, atrioventricular; BMI, body mass index; CHF, congestive heart failure; CV, cardiovascular; DBP, diastolic blood pressure; ETT, exercise tolerance test; LVEF, left ventricular ejection fraction; RT, radiation therapy; SBP, systolic blood pressure.
P-values were obtained from analysis of variance or χ2 test comparing across all categories.
LVEF data available for 1133 (69.4%) patients.
Cardiorespiratory fitness and mortality
During a median follow-up of 4.6 (IQR: 2.6–7.0) years, 411 deaths were observed: 50 CV deaths, 132 cancer deaths, and 229 from other causes. High CRF was associated with a significant reduction in the cumulative incidence of all-cause mortality at 5 and 10 years (Figure 2). For high CRF, the cumulative mortality rate at 5 years was 4% (95% CI: 3–6%) compared to 10% (95% CI: 8–14%) for intermediate CRF and 20% (95% CI: 17–25%) for low CRF (P < 0.001). At 10 years, the cumulative mortality was 13% (95% CI: 8–21%) for high CRF compared with 20% (95% CI: 15–27%) and 42% (95% CI: 36–49%) for intermediate and low CRF, respectively (P < 0.001).
Figure 2.
Probability of all-cause mortality (A), cardiovascular mortality (B), and cancer mortality (C) in cancer patients according to cardiorespiratory fitness tertiles achieved during exercise treadmill testing: low cardiorespiratory fitness (blue line), intermediate cardiorespiratory fitness (red line), and high cardiorespiratory fitness (green line).
All-cause mortality
For the primary endpoint of all-cause mortality, there was a significant inverse and graded relationship across increasing CRF tertiles (P < 0.001) (Figure 2). Compared with the lowest CRF tertile, the adjusted HR for all-cause mortality was 0.38 (95% CI: 0.28–0.52) for intermediate CRF and 0.17 (95% CI: 0.11–0.27) for high CRF (Table 2). A one-MET increment in CRF was associated with a 26% (95% CI: 0.70–0.79) relative reduction in the risk of all-cause mortality (Table 3).
Table 2.
Age-adjusted and multivariable-adjusted rate ratios of all-cause and cause-specific mortality according to tertiles of cardiorespiratory fitness
| METs |
||||
|---|---|---|---|---|
| Total | Low CRF | Intermediate CRF | High CRF | |
| All-cause mortality | ||||
| Number of events | 229 | 143 | 58 | 28 |
| Age-adjusted HR (95% CI) | Ref | 0.44 (0.32–0.60) | 0.26 (0.17–0.40) | |
| Multivariable-adjusted HR (95% CI)a | Ref | 0.38 (0.28–0.52) | 0.17 (0.11–0.27) | |
| Cardiovascular mortality | ||||
| Number of events | 50 | 32 | 10 | 8 |
| Age-adjusted HR (95% CI) | Ref | 0.39 (0.19–0.80) | 0.45 (0.19–1.04) | |
| Multivariable-adjusted HR (95% CI)a | Ref | 0.40 (0.19–0.86) | 0.41 (0.16–1.05) | |
| Cancer mortality | ||||
| Number of events | 132 | 77 | 37 | 18 |
| Age-adjusted HR (95% CI) | Ref | 0.47 (0.32–0.71) | 0.25 (0.14–0.43) | |
| Multivariable-adjusted HR (95% CI)a | Ref | 0.38 (0.26–0.60) | 0.16 (0.09–0.28) | |
Multivariable model adjusted for age, gender, race, body mass index, Morise risk score, atrioventricular nodal blockers, interval from cancer diagnosis to exercise treadmill test, and result of exercise treadmill test.
Table 3.
Risk of all-cause, cardiovascular, and cancer death for every one metabolic equivalent increase in cardiorespiratory fitness
| Age-adjusted hazard ratio per MET increase in exercise capacity (95% CI) | Adjusteda hazard ratio per MET increase in exercise capacity (95% CI) | |
|---|---|---|
| All-cause death | 0.79 (0.75–0.83) | 0.74 (0.70–0.79) |
| Cardiovascular deathb | 0.86 (0.76–0.98) | 0.86 (0.76–0.97) |
| Cancer deathb | 0.80 (0.74–0.85) | 0.75 (0.69–0.80) |
Adjusted for age, gender, race, body mass index, Morise risk score, AV nodal blockers, interval from cancer diagnosis to exercise treadmill test, and result of exercise treadmill test.
Risk of cardiovascular and cancer death was calculated using a competing risk model.
Cardiovascular and cancer mortality
The adjusted HRs for CV mortality were 0.40 (95% CI: 0.19–0.86) for intermediate CRF and 0.41 (95% CI: 0.16–1.05) for high CRF relative to the lowest CRF tertile. Similarly, the adjusted HRs for cancer mortality were 0.40 (95% CI: 0.26–0.60) for intermediate CRF and 0.16 (95% CI: 0.09–0.28) for high CRF relative to the lowest CRF tertile (Table 2). A one-MET increment in CRF was associated with a 14% (95% CI: 0.76–0.97) and 25% (95% CI: 0.69–0.80) relative reduction in the risk of CV and cancer death, respectively (Table 3).
Discussion
Our analysis of a diverse cohort with adult-onset cancer indicates that high post-diagnosis CRF is associated with substantial reductions in the risk of all-cause, CV, and cancer mortality after controlling for important clinical covariates. Collectively, these findings demonstrate the powerful prognostic importance of CRF in patients with cancer and identify CRF as a potential therapeutic target to improve outcomes following a cancer diagnosis.
Findings from this study are consistent with the wealth of data available in the general adult population as well as in those with overt clinical diagnoses (e.g. coronary heart disease, chronic obstructive pulmonary disease, heart failure) demonstrating that high CRF is associated with substantial reductions in mortality risk.15–18,25,27–29 In totality, the existing literature suggests that a one-MET increment in CRF is associated with a 12–25% decrease in mortality risk among both men and women. Majority of the studies examining the prognostic importance of CRF in apparently healthy adults have focused predominantly on overall mortality or CV mortality—however, it appears that high CRF is also protective against the primary incidence of various forms of cancer. For instance, midlife CRF was inversely associated with incidence of lung and colorectal cancer in 13 949 community-dwelling men without cancer, with each one MET increment associated with a 17% and 9% relative risk reduction, respectively.30
In the oncology setting, majority of the work to date has focused on whether self-reported physical activity/exercise following certain cancer diagnoses is associated with clinical outcomes. In brief, in a recent systematic review of 26 cohort studies with early-stage breast, colorectal, and prostate cancer patients, high levels of physical activity/exercise were associated with a pooled 37% reduction in cancer-specific mortality (pooled relative risk = 0.63; 95% CI: 0.54–0.73), compared to inactivity.31 Similarly, in a cohort study of 15 450 adult survivors of childhood cancer, there was a significant inverse relationship between self-reported exercise and all-cause mortality.32 While there are several advantages to self-reported methodology, particularly in large epidemiological studies, these methods have well-documented reliability and validity issues. In contrast, ETTs provide an objective assessment of physical activity/exercise via measurement of CRF and hence offer several advantages over self-reported methods. Unfortunately, limited clinical use of ETT in cancer patients has significantly hampered the ability to determine whether the powerful prognostic value of CRF extends to patients with cancer. Jones et al.33 found that high CRF, as evaluated by direct metabolic measurement (i.e. peak oxygen consumption), was inversely associated with the risk of death in patients with operable lung and metastatic breast cancer.34 At least two other studies report similar findings in haematological malignancies prior to stem cell transplantation.35,36 These studies, however, were small (n = 21–398), with limited follow-up duration, and consequently a small number of events. To this end, our analysis in ∼1600 patients with a relatively high number of events (411 deaths), showing that high CRF is associated with significant reductions in the risk of death, substantially extends the evidence base.
Against this background, our findings have two potential important clinical implications. First, current surveillance/screening guidelines for the prevention/detection of cancer treatment-associated CV dysfunction focus almost exclusively on CV risk factors, cardiac biomarkers, and echocardiographic parameters in cancer patients deemed at elevated risk of CV toxicity following treatment.37–39 Our findings indicate that ETT may provide complementary information to these assessments since it provides a comprehensive and integrated assessment of functional capacity not captured by current measures used in clinical practice. Indeed, in this study, 118 (7.2%) patients achieved <5 METs during ETT. Consequently, these patients have increased susceptibility to premature mortality, independent of traditional CV risk factors.25,27
Second, CRF is highly modifiable. Therefore, it is not only useful as a biomarker of long-term CV risk but is also an important therapeutic target. Aerobic exercise training is the most effective therapy to improve CRF in healthy individuals since it improves the reserve capacity of all oxygen transport organs, which together lead to favourable improvements in CV function. A recent meta-analysis by our group found that exercise training in patients with cancer was associated with a significant increase in CRF compared with usual care controls.40 Furthermore, among adult survivors of childhood cancer, a mean increase in self-reported exercise of 7.9 ± 4.4 MET-hours/week over an 8-year period was associated with a 40% reduction in all-cause mortality, relative to maintenance of low exercise.32 Indeed, a recent Scientific Statement from the American Heart Association proposes cardiac rehabilitation as a way to mitigate CV risk in select patients with cancer.41 Prospective studies are needed to clarify whether improvements in CRF confer survival benefit in various cancer cohorts.
Our study has several limitations. First, this was a single-centre study that evaluated a cohort of cancer patients referred for exercise testing for clinical reasons. Consequently, our cohort had an increased prevalence of CV risk factors as exemplified by an intermediate-risk Morise score. While this may have strengthened the association between reduced CRF and CV mortality relative to a non-referral cohort, our results agree with prior observations in non-cancer patients referred for ETT.25,28,42 Second, we selected patients with four different cancers that were highly represented in our database. A model that adjusted for age, sex, ETT result, and cancer type, did not change the association of categorical CRF with all-cause, CV, and cancer death (data not shown). Third, the median interval from cancer diagnosis to ETT was long and differed across CRF tertiles; however, associations between CRF and endpoints remained significant in multivariable analyses that included this parameter. Nonetheless, we acknowledge that multivariable analyses do not account for residual potential confounding of results by several unrecorded variables that influence CRF following cancer diagnosis, including cancer stage. Therefore, features of our study design may have introduced selection and survival biases that may limit the generalizability of our study findings to other cancer cohorts. In addition, exercise effort achieved may have been submaximal resulting in an under-estimate of CRF in some patients where exercise was terminated due to an exercise endpoint other than exhaustion.
In conclusion, our findings demonstrate that post-diagnosis CRF is a strong, independent predictor of all-cause, CV, and cancer mortality in patients with certain cancers. If these findings can be reproduced in multicentre studies involving other cancer cohorts, objective exercise testing after completion of cancer treatment may offer a reliable way to identify those at increased risk for adverse outcomes. The role of cardio-oncology rehabilitation to mitigate risk in patients with cancer needs further investigation.
Funding
This work was supported by the Linda Pollin Award and Goodman Scholar Award, Brigham and Women’s Hospital, Boston, MA granted to Dr John Groarke and through the Catherine Fitch Fund and Gelb Scholar Fund, Brigham and Women’s Hospital, Boston, MA supporting Dr Anju Nohria.
Conflict of interest: The authors have no conflict of interest relevant to this work to declare.
Contributor Information
John D Groarke, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA; Adult Survivorship Program, Department of Medical Oncology, Dana-Farber Cancer Institute, Brigham and Women’s Hospital, 450 Brookline Avenue, Boston, MA 02215, USA.
David L Payne, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA.
Brian Claggett, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA.
Mandeep R Mehra, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA.
Jingyi Gong, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA.
Jesse Caron, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA.
Syed S Mahmood, Department of Medicine, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA.
Jon Hainer, Noninvasive Cardiovascular Imaging Program, Department of Radiology, Brigham and Women’s Hospital, Boston, MA 02115, USA.
Tomas G Neilan, Cardio-Oncology Program, Division of Cardiology, Department of Medicine, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114-2696, USA.
Ann H Partridge, Adult Survivorship Program, Department of Medical Oncology, Dana-Farber Cancer Institute, Brigham and Women’s Hospital, 450 Brookline Avenue, Boston, MA 02215, USA.
Marcelo Di Carli, Noninvasive Cardiovascular Imaging Program, Department of Radiology, Brigham and Women’s Hospital, Boston, MA 02115, USA.
Lee W Jones, Department of Medicine, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA.
Anju Nohria, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA; Adult Survivorship Program, Department of Medical Oncology, Dana-Farber Cancer Institute, Brigham and Women’s Hospital, 450 Brookline Avenue, Boston, MA 02215, USA.
References
- 1. Verweij NM, Schiphorst AH, Pronk A, van den Bos F, Hamaker ME. Physical performance measures for predicting outcome in cancer patients: a systematic review. Acta Oncol 2016;55:1386 –1391. [DOI] [PubMed] [Google Scholar]
- 2. Tampellini M, Berruti A, Gerbino A, Buniva T, Torta M, Gorzegno G et al. Relationship between CA 15-3 serum levels and disease extent in predicting overall survival of breast cancer patients with newly diagnosed metastatic disease. Br J Cancer 1997;75:698 –702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Greenberg PA, Hortobagyi GN, Smith TL, Ziegler LD, Frye DK, Buzdar AU. Long-term follow-up of patients with complete remission following combination chemotherapy for metastatic breast cancer. J Clin Oncol 1996;14:2197 –2205. [DOI] [PubMed] [Google Scholar]
- 4. Alexandre J, Bleuzen P, Bonneterre J, Sutherland W, Misset JL, Guastalla J et al. Factors predicting for efficacy and safety of docetaxel in a compassionate-use cohort of 825 heavily pretreated advanced breast cancer patients. J Clin Oncol 2000;18:562 –573. [DOI] [PubMed] [Google Scholar]
- 5. Ryberg M, Nielsen D, Osterlind K, Skovsgaard T, Dombernowsky P. Prognostic factors and long-term survival in 585 patients with metastatic breast cancer treated with epirubicin-based chemotherapy. Ann Oncol 2001;12:81 –87. [DOI] [PubMed] [Google Scholar]
- 6. Parnes HL, Cirrincione C, Aisner J, Berry DA, Allen SL, Abrams J et al. Phase III study of cyclophosphamide, doxorubicin, and fluorouracil (CAF) plus leucovorin versus CAF for metastatic breast cancer: cancer and Leukemia Group B 9140. J Clin Oncol 2003;21:1819 –1824. [DOI] [PubMed] [Google Scholar]
- 7. Falkson G, Holcroft C, Gelman RS, Tormey DC, Wolter JM, Cummings FJ. Ten-year follow-up study of premenopausal women with metastatic breast cancer: an Eastern Cooperative Oncology Group study. J Clin Oncol 1995;13:1453 –1458. [DOI] [PubMed] [Google Scholar]
- 8. Chow R, Chiu N, Bruera E, Krishnan M, Chiu L, Lam H et al. Inter-rater reliability in performance status assessment among health care professionals: a systematic review. Ann Palliat Med 2016;5:83 –92. [DOI] [PubMed] [Google Scholar]
- 9. Hurria A, Gupta S, Zauderer M, Zuckerman EL, Cohen HJ, Muss H et al. Developing a cancer-specific geriatric assessment: a feasibility study. Cancer 2005;104:1998 –2005. [DOI] [PubMed] [Google Scholar]
- 10. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;56:M146 –M156. [DOI] [PubMed] [Google Scholar]
- 11. Sorror ML, Maris MB, Storb R, Baron F, Sandmaier BM, Maloney DG et al. Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood 2005;106:2912 –2919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Stenehjem JS, Smeland KB, Murbraech K, Holte H, Kvaloy S, Thorsen L et al. Cardiorespiratory fitness in long-term lymphoma survivors after high-dose chemotherapy with autologous stem cell transplantation. Br J Cancer 2016;115:178 –187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Jones LW, Hornsby WE, Goetzinger A, Forbes LM, Sherrard EL, Quist M et al. Prognostic significance of functional capacity and exercise behavior in patients with metastatic non-small cell lung cancer. Lung Cancer 2012;76:248 –252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Jones LW, Eves ND, Haykowsky M, Freedland SJ, Mackey JR. Exercise intolerance in cancer and the role of exercise therapy to reverse dysfunction. Lancet Oncol 2009;10:598 –605. [DOI] [PubMed] [Google Scholar]
- 15. Aaronson KD, Mancini DM. Is percentage of predicted maximal exercise oxygen consumption a better predictor of survival than peak exercise oxygen consumption for patients with severe heart failure? J Heart Lung Transplant 1995;14:981 –989. [PubMed] [Google Scholar]
- 16. Kavanagh T, Mertens DJ, Hamm LF, Beyene J, Kennedy J, Corey P et al. Prediction of long-term prognosis in 12 169 men referred for cardiac rehabilitation. Circulation 2002;106:666 –671. [DOI] [PubMed] [Google Scholar]
- 17. Kavanagh T, Mertens DJ, Hamm LF, Beyene J, Kennedy J, Corey P et al. Peak oxygen intake and cardiac mortality in women referred for cardiac rehabilitation. J Am Coll Cardiol 2003;42:2139 –2143. [DOI] [PubMed] [Google Scholar]
- 18. Kubozono T, Itoh H, Oikawa K, Tajima A, Maeda T, Aizawa T et al. Peak VO(2) is more potent than B-type natriuretic peptide as a prognostic parameter in cardiac patients. Circ J 2007;72:575 –581. [DOI] [PubMed] [Google Scholar]
- 19. Forman DE, Arena R, Boxer R, Dolansky MA, Eng JJ, Fleg JL et al. Prioritizing functional capacity as a principal end point for therapies oriented to older adults with cardiovascular disease: a scientific statement for healthcare professionals from the American Heart Association. Circulation 2017;135:e894 –e918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Groarke JD, Mahmood SS, Payne D, Ganatra S, Hainer J, Neilan TG et al. Case-control study of heart rate abnormalities across the breast cancer survivorship continuum. Cancer Med 2019;8:447 –454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Groarke JD, Tanguturi VK, Hainer J, Klein J, Moslehi JJ, Ng A et al. Abnormal exercise response in long-term survivors of Hodgkin lymphoma treated with thoracic irradiation: evidence of cardiac autonomic dysfunction and impact on outcomes. J Am Coll Cardiol 2015;65:573 –583. [DOI] [PubMed] [Google Scholar]
- 22. Bruce RA, Cooper MN, Gey GO, Fisher LD, Peterson DR. Variations in responses to maximal exercise in health and in cardiovascular disease. Angiology 1973;24:691 –702. [DOI] [PubMed] [Google Scholar]
- 23. Fletcher GF, Ades PA, Kligfield P, Arena R, Balady GJ, Bittner VA et al. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation 2013;128:873 –934. [DOI] [PubMed] [Google Scholar]
- 24. Gibbons RJ, Balady GJ, Beasley JW, Bricker JT, Duvernoy WF, Froelicher VF et al. ACC/AHA guidelines for exercise testing: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Exercise Testing). Circulation 1997;96:345–354. [DOI] [PubMed] [Google Scholar]
- 25. Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood JE. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med 2002;346:793 –801. [DOI] [PubMed] [Google Scholar]
- 26. Morise AP. Comparison of the Diamond-Forrester method and a new score to estimate the pretest probability of coronary disease before exercise testing. Am Heart J 1999;138:740 –745. [DOI] [PubMed] [Google Scholar]
- 27. Gulati M, Pandey DK, Arnsdorf MF, Lauderdale DS, Thisted RA, Wicklund RH et al. Exercise capacity and the risk of death in women: the St James Women Take Heart Project. Circulation 2003;108:1554 –1559. [DOI] [PubMed] [Google Scholar]
- 28. Roger VL, Jacobsen SJ, Pellikka PA, Miller TD, Bailey KR, Gersh BJ. Prognostic value of treadmill exercise testing: a population-based study in Olmsted County, Minnesota. Circulation 1998;98:2836 –2841. [DOI] [PubMed] [Google Scholar]
- 29. Cote CG, Pinto-Plata V, Kasprzyk K, Dordelly LJ, Celli BR. The 6-min walk distance, peak oxygen uptake, and mortality in COPD. Chest 2007;132:1778 –1785. [DOI] [PubMed] [Google Scholar]
- 30. Lakoski SG, Willis BL, Barlow CE, Leonard D, Gao A, Radford NB et al. Midlife cardiorespiratory fitness, incident cancer, and survival after cancer in men: the Cooper Center Longitudinal Study. JAMA Oncol 2015;1:231 –237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Friedenreich CM, Neilson HK, Farris MS, Courneya KS. Physical activity and cancer outcomes: a precision medicine approach. Clin Cancer Res 2016;22:4766 –4775. [DOI] [PubMed] [Google Scholar]
- 32. Scott JM, Li N, Liu Q, Yasui Y, Leisenring W, Nathan PC et al. Association of exercise with mortality in adult survivors of childhood cancer. JAMA Oncol 2018;4:1352 –1358. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Jones LW, Watson D, Herndon JE, Eves ND, Haithcock BE, Loewen G et al. Peak oxygen consumption and long-term all-cause mortality in nonsmall cell lung cancer. Cancer 2010;116:4825 –4832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Jones LW, Courneya KS, Mackey JR, Muss HB, Pituskin EN, Scott JM et al. Cardiopulmonary function and age-related decline across the breast cancer survivorship continuum. J Clin Oncol 2012;30:2530 –2537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Kelsey CR, Scott JM, Lane A, Schwitzer E, West MJ, Thomas S et al. Cardiopulmonary exercise testing prior to myeloablative allo-SCT: a feasibility study. Bone Marrow Transplant 2014;49:1330 –1336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Wood WA, Deal AM, Reeve BB, Abernethy AP, Basch E, Mitchell SA et al. Cardiopulmonary fitness in patients undergoing hematopoietic SCT: a pilot study. Bone Marrow Transplant 2013;48:1342 –1349. [DOI] [PubMed] [Google Scholar]
- 37. Armenian SH, Lacchetti C, Barac A, Carver J, Constine LS, Denduluri N et al. Prevention and monitoring of cardiac dysfunction in survivors of adult cancers: American Society of Clinical Oncology Clinical Practice Guideline. J Clin Oncol 2017;35:893 –911. [DOI] [PubMed] [Google Scholar]
- 38. Lancellotti P, Nkomo VT, Badano LP, Bergler J, Bogaert J, Davin L et al. Expert consensus for multi-modality imaging evaluation of cardiovascular complications of radiotherapy in adults: a report from the European Association of Cardiovascular Imaging and the American Society of Echocardiography. J Am Soc Echocardiogr 2013;26:1013 –1032. [DOI] [PubMed] [Google Scholar]
- 39. Zamorano JL, Lancellotti P, Rodriguez Munoz D, Aboyans V, Asteggiano R, Galderisi M et al. 2016 ESC Position Paper on cancer treatments and cardiovascular toxicity developed under the auspices of the ESC Committee for Practice Guidelines: the Task Force for cancer treatments and cardiovascular toxicity of the European Society of Cardiology (ESC. ). Eur Heart J 2016;37:2768 –2801. [DOI] [PubMed] [Google Scholar]
- 40. Scott JM, Zabor EC, Schwitzer E, Koelwyn GJ, Adams SC, Nilsen TS et al. Efficacy of exercise therapy on cardiorespiratory fitness in patients with cancer: a systematic review and meta-analysis. J Clin Oncol 2018;36:2297 –2305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Gilchrist SC, Barac A, Ades PA, Alfano CM, Franklin BA, Jones LW et al. ; On behalf of the American Heart Association Exercise, Cardiac Rehabilitation, and Secondary Prevention Committee of the Council on Clinical Cardiology; Council on Cardiovascular and Stroke Nursing; and Council on Peripheral Vascular Disease. Cardio-oncology rehabilitation to manage cardiovascular outcomes in cancer patients and survivors: a scientific statement from the American Heart Association. Circulation 2019;139:e997 –e1012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Mandsager K, Harb S, Cremer P, Phelan D, Nissen SE, Jaber W. Association of cardiorespiratory fitness with long-term mortality among adults undergoing exercise treadmill testing. JAMA Netw Open 2018;1:e183605. [DOI] [PMC free article] [PubMed] [Google Scholar]