Secondary Logo

Journal Logo


Exercise and the Tumor Microenvironment: Potential Therapeutic Implications

Wiggins, Jennifer M.1,2; Opoku-Acheampong, Alexander B.3; Baumfalk, Dryden R.3; Siemann, Dietmar W.1,2; Behnke, Bradley J.3,4

Author Information
Exercise and Sport Sciences Reviews: January 2018 - Volume 46 - Issue 1 - p 56-64
doi: 10.1249/JES.0000000000000137
  • Free
  • Journal Club

Key Points

  • Both diffusive and perfusive limitations to oxygen delivery are prevalent in solid tumors leading to hypoxic regions, a major factor in treatment resistance and an aggressive tumor microenvironment.
  • Compared with healthy tissue, most of the tumor vasculature differs structurally (incomplete smooth muscle coverage) and functionally (diminished vasoconstriction), which leads to an inability to modulate vascular resistance within a solid tumor.
  • Aerobic exercise can enhance perfusion of solid tumors and mitigate tumor hypoxia. Long-term exercise leads to a homogenous distribution of blood flow within tumors and increased basal oxygenation, suggesting a permanent exercise-induced modulation of the tumor microenvironment.
  • Emerging evidence suggests that exercise, in combination with conventional therapies, can lead to smaller tumors and may enhance prognosis for the patient.
  • Host-tissue vasoreactivity and tumor location are critical in the design and data interpretation from exercise and anticancer treatments in preclinical investigations.

Editor's note: Go online to view the Journal Club questions in the Supplemental Digital Content: see


Although physical activity is a fundamental component of our daily life, little is known about effects of exercise on the tumor microenvironment. The idea that exercise may prevent cancer development was first postulated early in the 20th century (1), and several elegant reviews have focused on physical activity and cancer risk (2), tumor progression (3), and cardiovascular disease in patients with cancer (4,5). However, an emerging area of research is the study of intratumoral events and systemic-tumor interactions occurring during exercise, as well as after exercise training, that may influence prognosis for the patient and treatment efficacy. As shown in Figure 1, inadequate vascular networks in solid tumors lead to regions of low oxygen tensions containing hypoxic tumor cells that have long been known to be resistant to anticancer therapy (6,8) and promote the adoption of an aggressive phenotype (9), tumor progression (10), and dissemination (6). This review focuses on recent research investigating systemic/tumor interactions that occur during exercise and how such interactions may affect molecular and cellular functions within the tumor microenvironment. We hypothesize that, due to the aberrant nature on the tumor vasculature and enhanced mean arterial pressure (MAP), aerobic exercise increases tumor blood flow, recruits previously nonperfused tumor blood vessels, and thereby augments blood-tumor O2 transport and diminishes tumor hypoxia. Further, due in part to a more homogenous distribution of tumor blood flow and enhanced tumor oxygenation, when combined with conventional anticancer treatments, aerobic exercise can significantly improve the outcomes for several types of cancers. Our hypothesis is illustrated in Figures 1B and C. Briefly, at rest, there is limited blood flow to a tumor, based upon cardiac output distribution, host-tissue blood flow, and vascular resistance within the tumor. However, during exercise, due to the dysfunctional vasculature of the tumor, with an increase in cardiac output and MAP, the increased perfusion pressure to the tumor (and the inability of tumor blood vessels to vasoconstrict) results in a significant increase in tumor blood flow and reduction in tumor hypoxia. This review focuses on the tumor microenvironment and its relation to prognosis for the patient, followed by support of our hypothesis focusing on the unique features of the tumor vasculature and how these can be exploited by exercise to manipulate tumor oxygen delivery acutely, and with chronic exercise training.

Figure 1
Figure 1:
A. Components of the tumor vasculature leading to intratumoral hypoxia, associated outcomes, and negative consequences to treatment and aggressiveness of the disease. Tumor hypoxia and blood flow at rest (B) and during exercise (C) and potential mechanisms for altered blood flow and enhanced oxygenation during exercise within the tumor. There are several mechanisms contributing to tumor hypoxia. Panel B illustrates the mechanisms of diffusion limited hypoxia, in which the spatial location of tumor cells is beyond the diffusion limit of O2, and perfusion limited hypoxia, in which temporal fluctuations in blood flow can result in blood vessels that are not actively supporting perfusion thus limiting O2 delivery. Noteworthy, diffusion-limited hypoxia reflects primarily chronic or long-duration (h·d−1) hypoxia, whereas perfusion-limited or acute hypoxia is of a transient nature (seconds to minutes). Both types of hypoxia are associated with treatment resistance and enhanced metastatic potential (6,7). (Panel A reprinted from (6). Copyright © 2015 Elsevier. Used with permission.)

The Tumor Microenvironment

A major contributor to the failure of conventional cancer therapies is the tumor microenvironment. Solid tumors cannot grow beyond ~2 mm3 (limit of nutrient passive diffusion (11,12)) without developing their own blood vessel network (13). However, due to the excessive nutritional demands of the expanding neoplastic mass, tumor vasculature is highly abnormal (Fig. 2A). Tumor-feeding arteries (15) and microvasculature (16) are characterized by immature, tortuous, porous, heterogeneous, and unorganized vessels, which, when coupled with the high interstitial pressure of tumors, results in poor blood flow, nutrient delivery, and limited response to vasomotor control (15). As reviewed in Siemann and Horsman (17), a critical nutrient that is affected is oxygen, and regions of hypoxia, typically defined as a partial pressure of oxygen (pO2) < 10 mm Hg, are common in most cancers. Two types of hypoxia occur in tumors. Cancer cells existing at the edge of the oxygen diffusion distance, typically ≥100 μm from a blood vessel, experience diffusion limited or chronic hypoxia. Acute hypoxia occurs when tumor cells are deprived of oxygen because blood vessels are transiently occluded and blood flow is episodic that can occur at intervals ranging from a few minutes to hours. The consequences of either type of hypoxia are significant (Fig. 1A) and lead to therapeutic barriers that create a significant hindrance to the control of cancers by conventional anticancer therapies (reviewed in (6)).

Figure 2
Figure 2:
A. Smooth muscle (shown in red via α-smooth muscle actin staining) morphology of arteries perforating healthy tissue and those from two types of tumors (mammary and lung). Note the circumferential arrangement of smooth muscle around endothelial cells (shown in green via CD31 staining) vessel wall in the normal healthy condition in contrast with incomplete coverage in arteries from the tumors, with smooth muscle projecting into the tumor parenchyma. B. Contractile responses in arterioles from a prostate tumor and the healthy prostate to the sympathetic mimetic norepinephrine (NE). C. Myogenic vasoconstriction to increasing intraluminal pressure in arterioles from the healthy prostate (black bars) and those of the prostate tumors (white bars). Note at higher intraluminal pressures, tumor arterioles have diminished myogenic tone (constriction). α-SMA, α-smooth muscle actin. (Panel A adapted from (14). Copyright © 2002 Elsevier. Used with permission.) (Panels B and C reprinted from (15). Copyright © 2014 Oxford University Press. Used with permission.)

The molecular consequences of tumor hypoxia are an enhanced expression of numerous genes associated with tumor progression and metastasis, as discussed in Hill and Chaudary (18). Hypoxia activates key transcriptional mediators, in particular the hypoxia-inducible factor (HIF) family. A complete review on HIF regulation and its role in tumor progression has been elegantly described by LaGory and Giaccia (19). Briefly, HIF is an O2 sensitive heterodimer of the subunits, α (α1, α2, or α3) and β1, which stabilize under low O2 conditions and translocate to the nucleus. HIF-1α typically responds to acute changes in oxygenation, whereas HIF-2α is induced under extended periods of hypoxia, with less known regarding the regulation of HIF-3α. When dimerized, these transcription factors activate many genes involved in an adaptive response to the hypoxic environment. Some of these changes include the activation of cell stress, metabolic reprogramming, epithelial mesenchymal transition, angiogenesis, stem-like characteristics as well as resistance to immune-mediated cell death.

Tumor-initiating cells, also referred to as cancer stem cells, are a subpopulation of tumor cells characterized by their ability to self-renew and give rise to functionally different subclones. SOX2, an embryonic stem cell marker involved in tumor cell pluripotency and sphere formation, has recently been associated with HIF-1α and HIF-2α expression in prostate tumors (20). This suggests hypoxia may be involved in tumor cell invasion and stemness in a time-dependent manner through the induction of SOX2.

Hypoxia is not only involved in the promotion of tumor cell aggressiveness but also induces stromal cell changes that enhance tumor progression. The stroma, largely composed of fibroblasts, extracellular matrix, endothelial cells, and infiltrating immune cells, is sensitive to changes in oxygenation. Key mediators in the modulation of immune cell function are HIF-1α and HIF-2, which have been observed to play opposing roles in the M1 versus M2 polarization of macrophages. Tumor-suppressive M1 macrophages appear to be associated with specific expression levels of HIF-1α, whereas tumor-promoting M2 macrophages are associated with HIF-2 (21). Hypoxia, mainly, but not exclusively, through HIF, supports the maintenance of a proangiogenic and immunosuppressive microenvironment by activating vascular endothelial growth factor (VEGF) expression, recruiting T-regulatory cells, and upregulating T-cell programmed death-ligand 1 (PD-L1) expression, which blunts the cytotoxic T-cell-mediated antitumor response. Paracrine signals activated by HIF promote an immunosuppressive environment, by recruiting T-regulatory cells, promoting M2 macrophage polarization, and increasing PD-L1 expression, which is associated with poor survival rates (19). Thus, regardless of HIF-1α association with M1 macrophage maturation, hypoxia is a strong driver of tumor progression and immune suppression.

Furthermore, hypoxia, a common feature of most solid tumors (22), has been correlated with local tumor control and indicated as an independent predictor of disease progression and negative clinical outcomes in a number of settings (9,10,23). In addition to negatively affecting therapy response, hypoxia can promote genomic instability, metabolic adaptations, angiogenesis, and metastasis (7,24,25).

Approaches to Modulate Tumor Oxygenation

Numerous approaches seeking to diminish tumor hypoxia by improving tissue oxygenation (26) have suffered from an inability to target tumors specifically, without inducing systemic responses. Many such approaches (vasodilators, hyperoxic gas breathing) can affect whole body hemodynamics (such as MAP, systemic vascular resistance (SVR), and cardiac output (Q)) and regional blood flow distribution, resulting in an unpredictable tumor blood flow response to the therapy. Given that MAP is tightly regulated in a small range, based on the equation MAP = Q × SVR, any change to either SVR or cardiac output will induce a reciprocal change in the other variable to maintain MAP (27). For example, vasopressors improve the tumor-to-normal tissue blood flow ratio (28) but also rapidly elevate SVR and will result in a compensatory decrease in Q to maintain MAP. Indeed, changes in tumor flow with such approaches are of short duration (e.g., <3 min with norepinephrine (NE) (28)). Conversely, if there is a decrease in SVR with vasodilatory therapy, there will be an increase in Q, but the reduced vascular resistance may shunt blood flow away from the tumor. Administration of vasodilators can reduce blood flow to a tumor (29). Furthermore, if dilator therapy is prolonged, there can be an increase in the blood stored within the periphery, reducing cardiac preload and diminishing the ability to maintain cardiac output (and MAP). Therefore, the physiological parameters involved in tumor perfusion at the systemic level can be manipulated within only a small range due to rigorous protection of MAP in the resting state. Consistent with our hypothesis, we propose that the regulatory mechanisms that maintain MAP and prevent current techniques from improving oxygenation in the tumor can be overcome by aerobic exercise.

Tumor Vascular Dysfunction, Blood Flow, and Oxygenation During Exercise

Optimal nutrient delivery to any tissue requires a coordinated balance between vasodilatory and vasoconstrictor influences. Mediators of vascular tone include, but are not limited to, the central nervous center (such as sympathetic release of the neurotransmitter NE), humoral stimuli (such as epinephrine), and local control (such as myogenic autoregulation). Key to the regulation of blood flow is a functional endothelium and smooth muscle of arteries and arterioles, with a circumferential organization around the vessel wall (Fig. 2A) to facilitate contraction or relaxation. The tumor vasculature often possesses poorly developed intimal and medial layers lacking in functional endothelium and smooth muscle, respectively (14,30). Figure 2A demonstrates the abnormal morphology of blood vessels in two types of tumors, including the incomplete coverage of the vessel wall with smooth muscle, as well as smooth muscle cells projecting into the tumor parenchyma (14). The pathophysiological consequences of these structural abnormalities include an impaired regulation of vascular tone and heterogeneous pattern of blood flow (31), contributing to local areas of hypoxia as discussed previously.

Normally, sympathetic activation results in the release of the neurotransmitter NE from nerve terminals, which then binds to alpha receptors on the smooth muscle to elicit a contraction (vasoconstriction). However, unlike most arteries, tumor vessels lack significant innervation (32) and functional smooth muscle (Fig. 2A). Furthermore, tumor-perforating arteries from preclinical orthotopic prostate cancer models (tumor implanted in the organ of origin) show little to no vasoconstriction in response to NE (Fig. 2B (15)) (or phenylephrine; data not shown) versus a robust constriction in arterioles perforating the healthy prostate. Therefore, resistance vessels (feed arteries of tumors) lack the neurogenic ability and functional smooth muscle requisite to vasoconstrict in response to sympathetic stimulation. The functional consequence of the lack of response to sympathetic agonists of tumor vessels would be a diminished vascular resistance response to NE. Indeed, Shankar et al. (28) demonstrated that, when exposed to noradrenaline (i.e., NE), tumor blood flow is not reduced as much as that of surrounding tissue (i.e., diminished constriction of the tumor vasculature). In addition to the blunted response to NE, tumor arteries demonstrate a diminished ability to maintain diameter with progressive increases in intraluminal pressure (i.e., reduced or absent myogenic vasoconstriction; Fig. 2C (15)), as would occur with exercise. Therefore, the inability to maintain diameter at higher blood pressures and the absent alpha-adrenergic vasoconstriction culminate in a compromised ability of the tumor vasculature to modulate vascular resistance. However, dysfunction in these pathways may allow for transient and prolonged modulation of tumor perfusion during exercise.

At rest, central venous pressure is low, with cardiac output directed primarily to compliant organs with a relatively higher vascular resistance in skeletal muscle. Under these conditions, arterial pressure is normal and tumor perfusion is low. Central venous pressure and cardiac output increase during exercise, sustained by the contracting muscle acting as a peripheral pump to drive venous blood flow back to the right ventricle (27). Coinciding with increased cardiac output, there is a redistribution of regional blood flow away from compliant tissue due, in part, to an enhanced sympathetic nerve activity. Specifically, the vasoconstriction of these "inactive" tissues (e.g., kidney and splanchnic organs (33)) increases local vascular resistance to either maintain or reduce blood flow to these tissues, thus "shunting" cardiac output to the contracting actively vasodilated tissues (such as the heart, respiratory, and skeletal muscles (34)). What are the ramifications of these systemic changes with exercise on the tumor microenvironment? Given the tumor vasculature cannot augment vascular resistance because of the impaired vasoconstriction to alpha-adrenergic agonists (Fig. 2B), a portion of the increased cardiac output is directed to the lower resistance of the tumor vasculature. In concert with the elevated cardiac output, during exercise there is an increase in MAP (i.e., the "pressor" response), and a resetting of the baroreflex (35) such that a higher blood pressure is tolerated. Therefore, in the face of an increasing pressure during exercise, the tumor arterioles, which have a diminished myogenic constriction (Fig. 2C), would have an increased diameter and, based upon Poiseuille law of fluid dynamics, mandate a greater blood flow to the tumor. Conceptually, the increased tumor blood flow can be explained based on the fundamentals of Ohm's law, which states current (I) equals the voltage differences (ΔV) divided by resistance (R). Specifically, Ohm's law can be applied to the vasculature based upon the following equation:

Where "current" is blood flow, the ΔV is the pressure difference (i.e., perfusion pressure) between P1 (arterial pressure) and P2 (venular pressure), and R is vascular resistance, determined primarily by feed arteries and arterioles. With exercise, there is a large increase in perfusion pressure and, in healthy nonactive tissue, this increased perfusion pressure is offset with an increased R to either maintain or decrease flow to the organ. In the tumor, the increased perfusion pressure in the face of an unchanged R could independently increase blood flow; however, the inability to maintain diameter in the face of an increased intraluminal pressure (Fig. 2C) would "open" (i.e., passive vasodilation) the tumor vasculature and drive down R, which also would increase blood flow. There is evidence that this series of events is precisely what occurs within a tumor during exercise, at least in preclinical studies in which tumor blood flow, perfusion, and oxygenation are measured during exercise.

A schematic of our general hypothesis (Fig. 1) depicts the systemic-tumor events occurring at rest (1B) and during exercise (1C) within a tumor. At rest, many tumor blood vessels do not support blood flow due, in part, to the high interstitial pressure within the tumor, which would lower transmural pressure and compress/collapse blood vessels. A consequence of the low blood flow and nonperfused vessels at rest is significant areas of the tumor demonstrating hypoxia, measured by the 2-nitroimidazole hypoxia-detection agent, EF5 in Figures 3A and B. As demonstrated by McCullough et al. (15) compared with resting conditions, during an acute bout of moderate-intensity (i.e., ~60%–70% of maximal aerobic capacity; V˙O2) aerobic exercise, prostate tumor blood flow increased ~200% compared with values at rest (Fig. 3C) in a preclinical orthotopic cancer model. However, it is not the total blood flow that determines oxygenation, but the distribution of perfusion.

Figure 3
Figure 3:
Prostate tumor-bearing rats were injected with 2-nitroimidazole, EF5, before a 60-min moderate-intensity run, during which EF5 was allowed to circulate and bind to hypoxic areas in the tumor. Animals were euthanized immediately after exercise and bound EF5 as a readout for tumor hypoxia was measured at rest and during exercise by immunofluorescence (A) and quantified (B). Tumor hypoxia was reduced by ~30% during exercise versus that measured at rest. C. Healthy prostate tissue and prostate tumor blood flow, measured by aortic infusion of radiolabeled microspheres, at rest and 5 min after the onset of exercise, reflecting the "steady-state" of exercise. Note the more than twofold increase in tumor blood flow during exercise. (Reprinted from (15). Copyright © 2014 Oxford University Press. Used with permission.)

The combination of enhanced flow and greater arterial pressure, with the latter increasing transmural pressure and potentially offsetting the high interstitial pressure in the tumor, likely leads to the perfusion of a significant number of tumor blood vessels that were not supporting blood flow at rest (15), suggesting a more homogenous spatial distribution of blood flow in the tumor. The increased blood flow and augmented functional tumor vasculature result in a significant reduction in tumor hypoxia during exercise (Figs. 3A, B). However, these data reflect a single bout of moderate-intensity exercise and there are no data regarding intratumoral blood flow or oxygenation with any other exercise intensity or duration. Still, the events occurring within a tumor during an acute bout of exercise provide the foundation for both predicting and understanding how aerobic exercise training, at least in the moderate-intensity domain, may alter the tumor microenvironment and hence conceivably affect treatment outcomes and prognosis.

Aerobic Exercise Training and Tumor Hypoxia

Unlike the paucity of tumor data collected during the course of exercise, more information is available that documents the effects of prolonged exercise training on the tumor microenvironment, from a variety of cancers in preclinical studies, with limited data from patients with cancer. The reader is directed to Ashcraft et al. (3) for a comprehensive literature review of the existing data. With respect to the hypoxic nature of the tumor microenvironment and tumor perfusion, aerobic exercise training is providing promising results, depending upon the location of the tumor and the responsiveness of the host tissue (36). As stated previously, during exercise there is a substantial increase in tumor blood flow, at least in orthotopic prostate cancer models (currently the only measurement of tumor blood flow during exercise). Increased blood flow through tissue provides a powerful stimulus for angiogenesis (37), and tissues exposed to chronic increases in blood flow (for example, with exercise training) demonstrate significant angiogenesis (37). The increased tumor blood flow (Fig. 3C) and increased functional vasculature during exercise (15) suggest there will be significant remodeling of the tumor vascular network with chronic exercise. Therefore, an increased vascular density of functional (perfused) vessels in the tumor after training is expected, which may improve O2 and anticancer therapy delivery (31). In preclinical models, the number of perfused vessels (basal) in orthotopic breast and prostate cancer models is increased after aerobic exercise training as shown through histological analysis after Hoechst-33342 intravenous injection (38,39). Indeed, a more homogenous tumor blood flow pattern after aerobic exercise training was confirmed using in vivo magnetic resonance imaging (MRI) perfusion mapping in a murine 4T1 orthotopic breast cancer model (40) (Fig. 4A). Importantly, the increased perfusion and vascularization with exercise training did not affect tumor size or growth rates (39), indicating no detrimental effect of such an intervention on gross tumor characteristics.

Figure 4
Figure 4:
A. In vivo magnetic resonance imaging (MRI) mapping of 4T1 mammary tumor perfusion (red/green) in a preclinical (orthotopic) model of breast cancer in sedentary and exercised animals. Exercise-trained animals showed greater and more homogenous perfusion in their mammary tumors compared with sedentary animals. B and C. Absolute and temporal profiles of microvascular pressure of oxygen (pO2) in prostate tumors from sedentary and exercise-trained groups. Note the doubling of tumor microvascular pO2 (B), measured via phosphorescence quenching, and loss of oscillations over time (C) in the group subjected to exercise training. D. Quantified by the hypoxia-detection agent EF5 in prostate tumors from sedentary and exercise-trained rats. After exercise training, tumor hypoxia in this orthotopic prostate cancer model was reduced significantly versus its sedentary counterparts. (Panel A from (40). Copyright © 2015 Oxford University Press. Used with permission.) (Panels B–D reprinted from (39). Copyright © 2013 The American Physiological Society. Used with permission.)

If exercise training increases the vascular volume of a tumor, it should enhance the diffusion gradient for transcapillary O2 flux and consequently improve tumor oxygenation. This was indeed the case in prostate tumor-bearing rats trained for 5 to 7 wk (39). Tumors of exercise-trained hosts showed significant increases in microvascular pO2 (Figs. 4B, C) and reductions in hypoxia (Fig. 4D) compared with tumors of sedentary counterparts. Not only did aerobic exercise training improve the microvascular pO2, reflecting an enhanced blood-tumor O2 driving force posttraining, but it also suppressed the oscillations in the microvascular pO2 profile that were observed in the tumor from the sedentary group (Fig. 4C (39)). The latter is important as the cyclic pO2 profile in tumors from sedentary subjects concurrent with the extremely low pO2 values (~3 mm Hg (39)) would likely create conditions of intermittent hypoxia, and enhanced reactive oxygen species generation (41) that are associated with tumorigenesis (42). Taken together, these findings indicate that in this preclinical orthotopic model of prostate cancer, aerobic exercise training can lead to a ~90% reduction in tumor hypoxia measured via EF5 (thus, reflecting primarily chronic hypoxia) (Fig. 4D (33)).

An unexpected finding from the preclinical studies on orthotopic models of breast and prostate cancer, which demonstrated either an enhancement of perfused vessels and a reduction in tumor hypoxia, is an increased expression of HIF-1α (43,44). As postulated by Jones et al. (44), the exercise-related increase in tumor HIF-1α expression was associated with an improved tumor perfusion (measured by MRI), thus reflecting an exercise-induced "stabilization" of HIF-1α. It is important to note that hypoxia is not requisite for activation and increased expression of HIF-1α. For example, HIF-1α expression is increased with cyclical mechanical stretch in vascular smooth muscle cells (45); thus, the increased stretch of tumor blood vessels associated with exercise also may provide a stimulus for the increased HIF-1α expression observed after training in these studies.

Despite these promising preclinical data, there is little information regarding the direct effects of exercise training on the tumor microenvironment of humans. This should not be surprising as withholding or prolonging the onset of adjuvant therapy to investigate the independent effects of exercise on the tumor microenvironment would raise significant ethical concerns. However, recent evidence suggests a downregulation of nuclear factor kappa light chain enhancer of activated B cells (NF-kB) in breast tumors of human subjects that were combining aerobic exercise training with chemotherapy versus chemotherapy alone (38). Thus, the work from Jones et al. (38) has demonstrated not only the feasibility, but potential benefits of aerobic exercise training in human patients with breast cancer undergoing neoadjuvant chemotherapy. Currently, there is only limited information for the effect of exercise on tumor characteristics for other types of cancer.

Exercise and the "Anti-Tumor" Environment

Aerobic exercise, specifically moderate-intensity exercise, may have the potential to improve therapeutic outcome. Not only by improving tumor perfusion and drug delivery, as proposed in our model, but also by engaging an antitumor systemic immune response. The innate immune system, our first response against foreign agents is composed, in part, of macrophages, neutrophils and natural killer (NK) cells.

In response to high intensity exercise, proinflammatory responses and immune cell function are reduced, dampening the immune response. As reviewed by Martin et al. (46), exercise intensity and infection risk tend to follow a "J curve," where the risk of upper respiratory tract infections depends on the exercise intensity. Conversely, moderate-intensity exercise seems to provide a protective effect against infections (17). An extensive review of the current literature on NK cytotoxicity by Zimmer et al. (47) uncovered that NK cell activation is dependent on the type of exercise, intensity, and frequency, but some exercise regimens can induce NK cell mobilization and activation, which may greatly impact the outcome. Indeed, a recent study in melanoma (48) demonstrated that voluntary exercise increased NK cell infiltration in the tumor that correlated with a decrease in tumor growth rate. In addition to NK cells, data published on the influence of exercise on macrophages (49) indicate that regular exercise can result in M1 macrophage polarization and better wound healing, thus promoting an antitumor response.

Considerations and Limitations

There are several considerations that must be acknowledged in both the design and interpretation of exercise oncology studies in humans and preclinical models. Key to predicting and understanding how a solid tumor responds to exercise is determining the host tissue and regional responses to exercise, with respect to cardiovascular control, exercise intensity and duration complicate this issue. The data presented in preclinical models reflect a mild-to-moderate intensity of exercise (~60% of V˙O2max). At this exercise intensity, blood flow to the reproductive organs does not change versus resting values (15), resulting in minimal regional or host-tissue reductions in perfusion determined by Hoechst-33342 staining procedure. In contrast, higher-intensity exercise with significant increases in plasma hormone concentrations (such as NE; Angiotensin II) may elevate host-tissue and regional vascular resistance, potentially reducing blood flow to a tumor, although this remains to be determined.

The tumor vasculature is intricately intertwined with the host tissue, such that host-tissue vascular responsiveness will undoubtedly influence tumor perfusion. Xenografts of tumor biopsies or tumor cells in rodents are common to determine tumorigenesis and the efficacy of therapeutic interventions. Ectopic transplantation of tumors or tumor cells injected subcutaneously are routine routes to grow and study cancers in preclinical models because of the ease of access and the minimal stress placed upon the animal. However, such ectopic tumors likely do not recapitulate tumor interactions with the host tissue of those grown in the organ of origin (orthotopic implantation). Given vascular function is organ and tissue specific, the tumor model and location are particularly important during exercise. This can be clearly demonstrated in preclinical models when the same tumor cell line is grown orthotopically versus ectopically (prostate tumor cells injected subcutaneously in the flank). In response to moderate-intensity exercise, the host tissue of an ectopic tumor (i.e., subcutaneous adipose tissue) demonstrates a reduction in blood flow, whereas there are no changes in perfusion of the host tissue of the orthotopic tumor (36). Consistent with altered host-tissue hemodynamics, blood flow responses to exercise in the ectopic are reduced and are increased in the orthotopic tumors (36). This clearly shows the importance of the host tissue and its interaction with the tumor during exercise. Thus, in determining perfusion responses to exercise (or other interventions that may impact local or systemic cardiovascular responses), tumor location is critical for the correct interpretation of results whether in preclinical models or in humans.

Most preclinical animal studies use voluntary versus forced exercise (3) to reduce the stress of the intervention on the animal(s) and induce an "exercised" phenotype in the animal. It should be noted that exercise in humans, particularly over 50%–60% of their aerobic capacity, induces the release of several stress hormones (such as cortisol). Thus, with most exercise interventions in humans and animals, stress is unavoidable, but can be controlled. The challenge with voluntary exercise is the quantification of parameters such as exercise intensity, which makes determining the mechanistic basis of exercise on the tumor microenvironment and translation difficult. As the reader can appreciate, there are significant differences, locally as well as systemically (i.e., blood pressure, humoral response, etc.), depending upon exercise modality (interval vs continuous), intensity (moderate/severe/exhaustive), and duration. Thus, detailed exercise prescription (vs obscure exercise recommendations such as kcal·d−1) is needed to translate specific mechanisms of cancer progression and changes in the tumor microenvironment in animal models, as well as to provide the personalized medicine patients with cancer deserve.


Exercise has been known to have a positive impact for the patient undergoing cancer treatment by mitigating therapy side effects. Furthermore, exercise is considered an integral component of disease recovery. Little is known regarding the effects of an acute bout or prolonged-training program of exercise on tumor physiology. We have provided evidence supporting our hypothesis that aerobic exercise may mitigate tumor hypoxia and alter the tumor microenvironment. Such changes will likely lead to a less aggressive tumor and potentially be more responsive to traditional (adjuvant) treatments. Outside of the tumor microenvironment, there are benefits of exercise at the systemic level such as increases in immune function that may create an "anti-tumor" environment. However, further studies are needed to understand 1) events occurring within a tumor during exercise in almost all cancer types, as current data reflect only prostate cancer, and 2) exercise parameters (i.e., intensity, duration, modality, etc.) associated with mechanistic changes within the tumor microenvironment to facilitate translation to the human patient, as there are only limited data currently in both preclinical models and human patients. Regardless, aerobic exercise therapy is seen as a "do no harm" approach that can impart significant benefits for the patient undergoing cancer treatment at both the systemic, and potentially tumor specific, levels and the adoption of a supervised exercise program should be considered a component of standard care with this disease. Notably, cancers such as breast, prostate, head and neck, and pancreas are prone to therapeutic failure due to the aberrant nature of the microenvironment and could potentially benefit from an exercise intervention.


Support for these projects was obtained from the American Cancer Society (RSG-14-150-1-CCE) to B.J.B.


1. Cherry T. A theory of cancer. Med. J. Aust. 1922; 1:425–38.
2. Moore SC, Lee IM, Weiderpass E, et al. Association of leisure-time physical activity with risk of 26 types of cancer in 1.44 million adults. JAMA Intern. Med. 2016; 176(6):816–25.
3. Ashcraft KA, Peace RM, Betof AS, Dewhirst MW, Jones LW. Efficacy and mechanisms of aerobic exercise on cancer initiation, progression, and metastasis: a critical systematic review of in vivo preclinical data. Cancer Res. 2016; 76(14):4032–50.
4. Scott JM, Adams SC, Koelwyn GJ, Jones LW. Cardiovascular late effects and exercise treatment in breast cancer: current evidence and future directions. Can. J. Cardiol. 2016; 32(7):881–90.
5. Scott JM, Koelwyn GJ, Hornsby WE, et al. Exercise therapy as treatment for cardiovascular and oncologic disease after a diagnosis of early-stage cancer. Semin. Oncol. 2013; 40(2):218–28.
6. Siemann DW, Horsman MR. Modulation of the tumor vasculature and oxygenation to improve therapy. Pharmacol. Ther. 2015; 153:107–24.
7. Cairns RA, Kalliomaki T, Hill RP. Acute (cyclic) hypoxia enhances spontaneous metastasis of KHT murine tumors. Cancer Res. 2001; 61(24): 8903–8.
8. Gray LH, Conger AD, Ebert M, Hornsey S, Scott OC. The concentration of oxygen dissolved in tissues at the time of irradiation as a factor in radiotherapy. Br. J. Radiol. 1953; 26(312):638–48.
9. Vaupel P. Hypoxia and aggressive tumor phenotype: implications for therapy and prognosis. Oncologist. 2008; 13(Suppl. 3):21–6.
10. Nelson DA, Tan TT, Rabson AB, Anderson D, Degenhardt K, White E. Hypoxia and defective apoptosis drive genomic instability and tumorigenesis. Genes Dev. 2004; 18(17):2095–107.
11. Blood CH, Zetter BR. Tumor interactions with the vasculature: angiogenesis and tumor metastasis. Biochim. Biophys. Acta. 1990; 1032(1):89–118.
12. Gimbrone MA Jr, Leapman SB, Cotran RS, Folkman J. Tumor dormancy in vivo by prevention of neovascularization. J. Exp. Med. 1972; 136(2):261–76.
13. Raica M, Cimpean AM, Ribatti D. Angiogenesis in pre-malignant conditions. Eur. J. Cancer. 2009; 45(11):1924–34.
14. Morikawa S, Baluk P, Kaidoh T, Haskell A, Jain RK, McDonald DM. Abnormalities in pericytes on blood vessels and endothelial sprouts in tumors. Am. J. Pathol. 2002; 160(3):985–1000.
15. McCullough DJ, Stabley JN, Siemann DW, Behnke BJ. Modulation of blood flow, hypoxia, and vascular function in orthotopic prostate tumors during exercise. J. Natl. Cancer Inst. 2014; 106(4):dju036.
16. Andrade SP, Bakhle YS, Hart I, Piper PJ. Effects of tumour cells on angiogenesis and vasoconstrictor responses in sponge implants in mice. Br. J. Cancer. 1992; 66(5):821–6.
17. Lowder T, Padgett DA, Woods JA. Moderate exercise protects mice from death due to influenza virus. Brain Behav. Immun. 2005; 19(5):377–80.
18. Hill RP, Chaudary N. Influence of hypoxia on metastatic spread. In: Tumor Microenvironment, edited by Siemann DW, editor. New York, John Wiley & Sons, 2011, p. 311–328.
19. LaGory EL, Giaccia AJ. The ever-expanding role of HIF in tumour and stromal biology. Nat. Cell Biol. 2016; 18(4):356–65.
20. Bae KM, Dai Y, Vieweg J, Siemann DW. Hypoxia regulates SOX2 expression to promote prostate cancer cell invasion and sphere formation. Am. J. Cancer Res. 2016; 6(5):1078–88.
21. Fang HY, Hughes R, Murdoch C, et al. Hypoxia-inducible factors 1 and 2 are important transcriptional effectors in primary macrophages experiencing hypoxia. Blood. 2009; 114(4):844–59.
22. Vaupel P, Mayer A. Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev. 2007; 26(2):225–39.
23. Movsas B, Chapman JD, Hanlon AL, et al. Hypoxic prostate/muscle pO2 ratio predicts for biochemical failure in patients with prostate cancer: preliminary findings. Urology. 2002; 60(4):634–9.
24. Graeber TG, Osmanian C, Jacks T, et al. Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature. 1996; 379(6560):88–91.
25. Dewhirst MW, Cao Y, Moeller B. Cycling hypoxia and free radicals regulate angiogenesis and radiotherapy response. Nat. Rev. Cancer. 2008; 8(6):425–37.
26. Jordan BF, Sonveaux P. Targeting tumor perfusion and oxygenation to improve the outcome of anticancer therapy. Front. Pharmacol. 2012; 3:94.
27. Rowell LB. Human Cardiovascular Control. New York: Oxford University Press; 1993.
28. Shankar A, Loizidou M, Burnstock G, Taylor I. Noradrenaline improves the tumour to normal blood flow ratio and drug delivery in a model of liver metastases. Br. J. Surg. 1999; 86(4):453–7.
29. Zlotecki RA, Baxter LT, Boucher Y, Jain RK. Pharmacologic modification of tumor blood flow and interstitial fluid pressure in a human tumor xenograft: network analysis and mechanistic interpretation. Microvasc. Res. 1995; 50(3):429–43.
30. Baluk P, Morikawa S, Haskell A, Mancuso M, McDonald DM. Abnormalities of basement membrane on blood vessels and endothelial sprouts in tumors. Am. J. Pathol. 2003; 163(5):1801–15.
31. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science. 2005; 307(5706):58–62.
32. Ashraf S, Loizidou M, Crowe R, Turmaine M, Taylor I, Burnstock G. Blood vessels in liver metastases from both sarcoma and carcinoma lack perivascular innervation and smooth muscle cells. Clin. Exp. Metastasis. 1997; 15(5):484–98.
33. Musch TI, Friedman DB, Pitetti KH, et al. Regional distribution of blood flow of dogs during graded dynamic exercise. J. Appl. Physiol. (1985). 1987; 63(6):2269–77.
34. Delp MD, Laughlin MH. Regulation of skeletal muscle perfusion during exercise. Acta Physiol. Scand. 1998; 162(3):411–9.
35. Raven PB, Fadel PJ, Ogoh S. Arterial baroreflex resetting during exercise: a current perspective. Exp. Physiol. 2006; 91(1):37–49.
36. Garcia E, Becker VG, McCullough DJ, et al. Blood flow responses to mild-intensity exercise in ectopic vs. orthotopic prostate tumors; dependence upon host tissue hemodynamics and vascular reactivity. J. Appl. Physiol. (1985). 2016; 121(1):15–24.
37. Hudlicka O, Brown M, Egginton S. Angiogenesis in skeletal and cardiac muscle. Physiol. Rev. 1992; 72(2):369–417.
38. Jones LW, Fels DR, West M, et al. Modulation of circulating angiogenic factors and tumor biology by aerobic training in breast cancer patients receiving neoadjuvant chemotherapy. Cancer Prev. Res. (Phila.). 2013; 6(9):925–37.
39. McCullough DJ, Nguyen LM, Siemann DW, Behnke BJ. Effects of exercise training on tumor hypoxia and vascular function in the rodent preclinical orthotopic prostate cancer model. J. Appl. Physiol. (1985). 2013; 115(12): 1846–54.
40. Betof AS, Lascola CD, Weitzel D, et al. Modulation of murine breast tumor vascularity, hypoxia and chemotherapeutic response by exercise. J. Natl. Cancer Inst. 2015; 107(5):1–5.
41. Dawson TL, Gores GJ, Nieminen AL, Herman B, Lemasters JJ. Mitochondria as a source of reactive oxygen species during reductive stress in rat hepatocytes. Am. J. Physiol. 1993; 264(4 Pt 1):C961–7.
42. Cook JA, Gius D, Wink DA, Krishna MC, Russo A, Mitchell JB. Oxidative stress, redox, and the tumor microenvironment. Semin. Radiat. Oncol. 2004; 14(3):259–66.
43. Jones LW, Viglianti BL, Tashjian JA, et al. Effect of aerobic exercise on tumor physiology in an animal model of human breast cancer. J. Appl. Physiol. (1985). 2010; 108(2):343–8.
44. Jones LW, Antonelli J, Masko EM, et al. Exercise modulation of the host-tumor interaction in an orthotopic model of murine prostate cancer. J. Appl. Physiol. (1985). 2012; 113(2):263–72.
45. Chang H, Shyu KG, Wang BW, Kuan P. Regulation of hypoxia-inducible factor-1alpha by cyclical mechanical stretch in rat vascular smooth muscle cells. Clin. Sci. (Lond.). 2003; 105(4):447–56.
46. Martin SA, Pence BD, Woods JA. Exercise and respiratory tract viral infections. Exerc. Sport Sci. Rev. 2009; 37(4):157–64.
47. Zimmer P, Schenk A, Kieven M, et al. Exercise induced alterations in NK-cell cytotoxicity — methodological issues and future perspectives. Exerc. Immunol. Rev. 2017; 23:66–81.
48. Pedersen L, Idorn M, Olofsson GH, et al. Voluntary running suppresses tumor growth through epinephrine- and IL-6-dependent NK cell mobilization and redistribution. Cell Metab. 2016; 23(3):554–62.
49. Goh J, Ladiges WC. Exercise enhances wound healing and prevents cancer progression during aging by targeting macrophage polarity. Mech. Ageing Dev. 2014; 139:41–8.

cancer; hypoxia; vascular control; oncology; oxygen delivery

Supplemental Digital Content

Copyright © 2017 by the American College of Sports Medicine