Volume 94, Issue 2 p. 539-551
Original Article
Free Access

Physical exercise and immune system function in cancer survivors

A comprehensive review and future directions

Adrian S. Fairey B.A.

Adrian S. Fairey B.A.

Center for Health Promotion Studies, University of Alberta, Edmonton, Canada

Search for more papers by this author
Kerry S. Courneya Ph.D.

Corresponding Author

Kerry S. Courneya Ph.D.

Faculty of Physical Education, University of Alberta, Edmonton, Canada

Fax: (780) 492-2364

Faculty of Physical Education, University of Alberta, E-424 Van Vliet Center, Edmonton, Alberta, T6G 2H9, Canada===Search for more papers by this author
Catherine J. Field Ph.D., R.D.

Catherine J. Field Ph.D., R.D.

Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Canada

Search for more papers by this author
John R. Mackey M.D.

John R. Mackey M.D.

Department of Oncology, University of Alberta, Edmonton, Canada

Search for more papers by this author
First published: 18 January 2002
Citations: 106

Abstract

BACKGROUND

There are a limited number of interventions for cancer survivors following completion of primary therapy that might reduce the risk of cancer recurrence and/or secondary malignancies and increase survival times. It has been proposed that physical exercise may be beneficial by enhancing the anticancer immune system response. The purpose of the current article is to: 1) briefly describe the immune system response to tumors, 2) discuss the impact of anticancer therapy on immune system function in cancer survivors, 3) provide a systematic and comprehensive review of the extant literature examining physical exercise and immune system function in cancer survivors, and 4) offer a critical analysis of this literature and outline directions for future research.

METHODS

A comprehensive literature search up to March 2001 identified empirical articles that examined the effects of physical exercise training on immune system function in cancer survivors from CD-ROM database searches and manual searches.

RESULTS

To the authors' knowledge, six empirical studies published between 1994 and 2000 have examined physical exercise and immune system function in cancer survivors. Overall, four out of six studies reported statistically significant improvements in a number of cancer-related immune system components as a result of exercise. However, there are several limitations that must be considered when interpreting the findings of these studies. These limitations involve the samples, designs, physical exercise interventions, physical fitness assessments, and immunologic assessments.

CONCLUSIONS

Additional research is needed to determine if physical exercise in cancer survivors may reduce the risk of cancer recurrence and secondary malignancies and increase survival times. Cancer 2002;94:539–51. © 2002 American Cancer Society.

Cancer is a major health problem in the United States. In 2000, an estimated 1.2 million Americans will be diagnosed with some form of invasive cancer.1 In addition, cancer will be the second leading cause of death in the U.S. in 2000, accounting for over 552,000 deaths.1 At present, Americans have about a 41% lifetime probability of developing cancer and about a 25% lifetime probability of dying from cancer.1 Improved detection and treatment of cancer have resulted in increased survival rates over the past few decades. Nevertheless, the five-year relative survival rate (adjusted for normal life expectancy) for all cancers combined is only 59%.1 Moreover, many other cancer survivors will be at a particularly high risk of developing a recurrence of their disease and/or a secondary malignancy.2 Therefore, interventions that can reduce the likelihood of a cancer recurrence and/or secondary malignancy and increase survival are needed.

Currently, there are a limited number of interventions for cancer survivors following completion of primary therapy that might reduce the risk of recurrence and/or secondary malignancies and increase survival times.3 Adjuvant drug treatment can be given to cancer survivors following surgical removal of their primary tumor and/or radiotherapy in an attempt to control clinically occult micrometastases.3 For hormone-sensitive breast and prostate carcinoma, adjuvant hormonal treatment can reduce the risk of both recurrence and death. For example, when administered daily for five years following surgical therapy to women with estrogen responsive breast carcinoma, tamoxifen therapy reduces the risk of recurrence by 35%, and the risk of death by 25%.4 Adjuvant chemotherapy, given after resection of breast carcinoma, colorectal carcinoma, gastric carcinoma, and osteosarcoma, among others, can improve disease-free survival and overall survival.3 However, adjuvant drug treatment frequently fails to control micrometastases, and survivors with other types of cancer lack proven posttreatment options for reducing their risk of disease recurrence and/or secondary malignancies. Consequently, research into other interventions designed to improve survival is warranted.

Physical exercise is a particularly attractive option because of its other known health benefits. Moreover, it has been proposed that physical exercise may positively influence one or more biologic systems important in anticancer defense. Recently, exercise-induced alterations in cancer-related immune system components have received increased research attention. The working theory in the field of exercise immunology is the Inverted J Hypothesis (Fig. 1).5 This hypothesis suggests that enhanced immune system function and reduced susceptibility to cancer occurs with regular moderate exercise. In contrast, repeated bouts of exhaustive exercise may lead to suppressed immune system function and elevated susceptibility to cancer.5 Based on this rationale, research examining the effects of physical exercise training on immune system function in cancer survivors has begun to emerge. However, to our knowledge, no systematic review of this literature has been conducted. The purpose of the current article is to: 1) briefly describe the immune system response to tumors, 2) discuss the impact of conventional anticancer therapy on immune system function in cancer survivors, 3) provide a systematic and comprehensive review of the extant literature examining physical exercise and immune system function in cancer survivors, and 4) offer a critical analysis of this literature and outline directions for future research.

Details are in the caption following the image

The Inverted J Hypothesis. The Inverted J Hypothesis suggests that enhanced immune system function and reduced susceptibility to cancer occurs with regular moderate exercise, whereas repeated bouts of exhaustive exercise may lead to suppressed immune system function and elevated susceptibility to cancer. Reprinted with permission from Med Sci Sports Exerc 1999;31:57-66.

Immune System Response to Tumors

The immune system protects against destructive forces either from outside the body (e.g., bacteria, viruses, and parasites) or from within (e.g., malignant and autoreactive cells). It comprises two functional divisions that work together in a coordinated manner. The innate immune system consists of cellular components, soluble factors, physical barriers, and the reticuloendothelial system.6 It provides a first line of defense against foreign pathogens while an acquired immune response is activated.7 The acquired immune system produces a specific reaction and immunologic memory to each pathogen and comprises cellular components and soluble factors.6 The innate and acquired immune systems are illustrated in Figures 2 and 3, respectively.

Details are in the caption following the image

The innate immune system. The innate immune system is composed of cellular components, soluble factors, physical defenses, and the reticuloendothelial system, all of which does not depend on previous exposure to a particular antigen. The cellular components of the innate immune system include natural killer cells and phagocytes (macrophages and neutrophils). The soluble factors include cytokines, complement, and acute phase proteins. Physical barriers include the skin and mucous membranes. The reticuloendothelial system comprises phagocytic cells that have a primary role in alleviating blood-borne infections. Adapted from: Glodsby RA, Kindt TJ, Osborne BA. Kuby Immunology. 4th ed. New York: W.H. Freeman, 2000. non-MHC: nonmajor histocompatibility complex.

Details are in the caption following the image

The acquired immune system. The acquired immune system is composed of both a cell mediated and a humoral branch. The cell-mediated component consists of two types of T lymphocytes: CD4+ (helper/inducer) and CD8+ (cytotoxic/suppressor). CD4+ cells are important in both the cell-mediated and humoral immune responses. The CD4+ subset is further divided into Th1 and Th2 immune cells based on the cytokines they produce and secrete. Th1 cells produce IL-2 and IFN-γ, whereas Th2 cells produce IL-4, IL-5, IL-10, and IL-13. Cytotoxic T cells (CD8+) directly kill target cells by recognizing antigen in association with the MHC Class I that is present on every cell in the human body. B cells are the major effector cells of the humoral system. They are activated to become antibody-secreting plasma cells. Adapted from: Glodsby RA, Kindt TJ, Osborne BA. Kuby Immunology. 4th ed. New York: W.H. Freeman, 2000. MHC: major histocompatibility.

The theory of immune surveillance asserts that many potential cancers might be eradicated prior to their clinical detection if they could be recognized and destroyed by acquired and/or innate immune system components.8, 9 According to this theory, tumors arise only if cancer cells are able to escape immune surveillance. Transformed cells are able to avoid destruction by the immune system either by reducing their expression of tumor antigens or by an impaired immune response to these cells.6 Support for this theory can be found in research that has shown that the immune system may have a particularly integral role in malignancies of viral origin.10 In particular, researchers have demonstrated an increased incidence of specific cancers in immunosuppressed patients with AIDS. These cancers include non-Hodgkin lymphoma, Kaposi sarcoma, anal carcinoma, and cervical carcinoma.11 Other findings, however, seem to challenge the tenets of the immune surveillance theory. For example, although individuals receiving immunosuppressive drugs show an increased incidence of immune system cancers, they do not show an increased incidence of other common cancers such as lung, breast, and colon carcinoma.6 Although the immune surveillance theory remains controversial, it is clear that both acquired and innate immune system components are able to produce an anticancer response to tumor cells.6

The acquired immune system must recognize tumor antigens in order to mount an anticancer T cell response. Tumor antigens located on tumor cells are either tumor-specific transplantation antigens (TSTA) or tumor-associated transplantation antigens (TATA).6 TSTA are unique to tumor cells and absent on normal cells. In contrast, TATA are not unique to tumor cells in that they are expressed on normal cells during fetal development, but are usually not expressed in adults.6 Tumor antigens recognized by human T cells can be classified as one of four types: 1) antigens encoded by genes explicitly expressed by tumors; 2) antigens encoded by deviate forms of normal genes that have been changed by mutation; 3) antigens typically expressed at certain stages of differentiation or only by certain differentiation lineages; and 4) antigens that are expressed in excess by certain tumors.6 Although some human cancers have been shown to be nonimmunogenic, research is accumulating to suggest that a specific immune response may still be an important defense against these tumor cells.6, 9, 12, 13 Human tumor antigens that have been associated with the anticancer immune system response are presented in Table 1.

Table 1. Human Tumor Antigens Associated with the Antitumor Immune Response14
Tumor rejection antigen Cancer Immune system recognition
Bcr/Abl Chronic myelogenous leukemia T cells
Mutated p53 Various T cells
Immunoglobulin idiotype B cell lymphoma Antibodies and T cells
Mutated Ras Adenocarcinomas T cells
MAGE 1 Melanoma T cells
MAGE 3 Melanoma T cells
MART 1 Melanoma T cells
Tyrosinase Melanoma T cells
Gp100 Melanoma T cells
MUC 1 Breast, pancreas T cells
  • Adapted From: Berinstein NL. Biological Therapy of Cancer. In: Tannock IF, Hill RP. The Basic Science of Oncology 3rd ed. New York: McGraw-Hill, 1998:420–442.

The innate immune system is also able to mount an anticancer response against tumor cells. Unlike the acquired immune system response, however, innate components can eliminate neoplastic cells without prior exposure to a particular antigen.6 Evidence that innate immune system reactions to tumor cells are important in host defense can be found in studies that have examined their immunologic role in the maintenance of cancer quiescence during remission.15 Animal research has also shown that enhanced tumor clearance is associated with high levels of activity by innate immune components.16, 17 Cellular innate immune system components involved in anticancer defense include natural killer (NK) cells, macrophages, and neutrophils. Soluble factors of the innate immune system implicated in cancer defense include C-reactive protein, interleukin 1 (IL-1), tumor necrosis factor (TNF), interferon (IFN), and oncostatin M.6 A brief review of the cellular components and soluble factors of the immune system that are involved in the antitumor response is provided in Tables 2 and 3.

Table 2. Cellular Immune System Components Involved in Cancer Defense6, 67
Component Anticancer function Mechanism(s) of action
Cytotoxic T lymphocytes Destroy tumor cells by apoptosis Granule-mediated exocytosis; Fas-Fas ligand interactions
Natural killer cells Destroy tumor cells in absence of MHC I and II antigen expression on target cells by apoptosis Opposing-signals model
Macrophages Destroy tumor cells Antibody-dependent cell-mediated cytotoxicity
Involved in antigen-presenting
Source of IL-1 and TNF
Neutrophils Destroy tumor cells through the production of peroxides and free radicals Antibody-dependent cell-mediated cytotoxicity
  • MHC: major histocompatibility complex; IL-1: interleukin 1; TNF: tumor necrosis factor.
  • Adapted from: Glodsby RA, Kindt TJ, Osborne BA. Kuby Immunology 4th ed. New York: W.H. Freeman, 2000 and Motyka B, Korbutt G, Pinkoski MJ, Heibein JA, Caputo A, Hobman M, et al. Mannose 6-phosphate/insulin-like growth factor II receptor is a death receptor for granzyme B during cytotoxic T cell-induced apoptosis. Cell 2000;103:491–500.
Table 3. Soluble Factors of the Immune System Involved in Cancer Defense6
Component Source Target cells Anticancer function
Acute phase proteins
 C-reactive protein Liver Tumor cells Render tumor cells more susceptible to phagocytosis by monocytes
Cytokines
 Interleukin-1 Monocytes, macrophages, B cells T cells Co-stimulates activation
B cells Promotes maturation and clonal expansion
NK cells Enhances cytotoxic activity
Macrophages/neutrophils Chemotactically attracts
 Tumor necrosis factor-α Macrophages Tumor cells Cytotoxic effects
 Tumor necrosis factor-β T cells Tumor cells Cytotoxic effects
Macrophages/neutrophils Enhances phagocytotic activity
 Interferon-α Leukocytes Tumor cells Inhibits tumor cell growth
NK cells Activates
Macrophages Activates
 Interferon-γ T cells, NK cells Macrophages Enhances cytotoxic activity
 Oncostatin M Macrophages, T cells Tumor cells Inhibits tumor cell growth
  • NK: natural killer.
  • Adapted from: Glodsby RA, Kindt TJ, Osborne BA. Kuby Immunology 4th ed. New York: W.H. Freeman, 2000.

Anticancer Therapy and the Immune System in Cancer Survivors

Cancer treatments have been shown to have dramatic effects on several immune system components in cancer survivors. In general, although the results are not entirely consistent, cancer therapies tend to be immunosuppressive. For example, Head et al. reported that breast carcinoma survivors who received chemotherapy demonstrated significantly impaired T and B cell response to mitogenic stimulation.18 In addition, Haku et al. reported that platinum-containing systemic chemotherapy significantly reduced the number of alveolar macrophages in survivors with lung carcinoma.19 Fludarabine, a nucleoside drug used in hematologic cancers, produces a profound and persistent depletion in T cell (CD4+) populations.20, 21 Furthermore, Stages I and II breast carcinoma survivors who received cyclic chemotherapy (CMF-cyclophosphamide, methotrexate, and 5-fluorouracil) have been shown to depress total lymphocyte counts as well as decrease T cells (CD4+, CD8+), B cells (CD19+), NK cells (CD3–CD16+CD56+), and activated T cells.22 Similarly, Sewell et al. showed that breast carcinoma survivors who received cyclic CMF had decreased numbers of T cells (CD4+) and B cells (CD19+), as well as reduced functional capacity in NK cells and lymphokine-activated killer (LAK) cells that persisted for up to six months following cessation of therapy.23

Studies indicate that radiotherapy can also depress several components of the immune system in cancer survivors. In particular, radiotherapy has been shown to cause significant reductions in both NK cell cytolytic activity in breast carcinoma survivors24 and the number of NK cells (CD3–CD16+CD56+) in colorectal carcinoma survivors.25 Similarly, Garzetti et al. reported that Stages I and II endometrial carcinoma survivors who received radiotherapy demonstrated a significant reduction in NK cell cytolytic activity.26 Squamous cell lung carcinoma survivors who received radiotherapy have been shown to have significant reductions in the total number of lymphocyte, T cells (CD4+, CD8+), and T cell proliferation in response to mitogenic stimulation.27 Furthermore, radiotherapy has been shown to reduce total lymphocyte counts, T cells (CD4+, CD8+), T cell (CD4+/CD8+) ratios, and T and B cell proliferation responses to mitogenic stimulation in survivors with squamous cell carcinoma of the oral cavity.28

Surgery is another cancer treatment modality that has been shown to suppress immune system function. More specifically, immune system components that have been shown to be impaired after surgical interventions include monocyte phagocytosis, antigen presentation and superoxide release, B cell immunoglobulin production, T cell response to mitogen stimulation, and IL-2 production.29 In addition, Uchida et al. reported that breast carcinoma survivors who received modified radical mastectomies had significantly reduced NK cell cytolytic activity that remained for more than two weeks following surgery.30 The ability to generate cells with LAK activity has also been shown to be reduced in cancer survivors who received surgery for gastric and colonic carcinomas.31-33 Notably, the suppressed generation of LAK cells was directly associated with a concomitant reduction in the number of NK cells (CD56+).33

Presently, however, there is limited research linking changes in immune system function during or after anticancer therapy to important cancer outcomes such as complications and/or the risk of recurrence. Nevertheless, preliminary studies indicate that the immunosuppressive effects of primary therapy may indeed be relevant. For instance, Head et al. showed that the magnitude of the decrease in the number of neutrophils and lymphocytes that occurred during chemotherapy was associated with disease relapse.18 In addition, Marana et al. found that survivors with advanced cervical carcinoma who received cisplatinum and bleomycin chemotherapy had reduced lymphocyte proliferation responses that correlated with prognosis and survival rates.34 Furthermore, McMillan et al. reported that reduced T cells (CD4+) occurred prior to detectable recurrence of colorectal carcinoma in survivors who had undergone surgery, which highlights the importance of impaired immunity in tumor recurrence.35 Although further research is needed on this topic, there is preliminary evidence to suggest that the decreased immune system response associated with conventional therapy may put cancer survivors at greater risk for disease recurrence and/or secondary malignancies.

Physical Exercise and Immune System Function in Cancer Survivors

A comprehensive literature search up to March 2001 identified empirical articles that examined the effects of physical exercise training on immune system function in cancer survivors from two principal sources: computer searches and manual searches. The computer searches utilized the CD-ROM databases of CancerLit, Dissertation Abstracts International, Embase, MedLine, PubMed, and SPORT Discus. Key terms that were combined and searched included physical activity, exercise, fitness, sport, cancer, tumor, neoplasm, carcinoma, leukemia, lymphoma, myeloma, sarcoma, immune system, immunity, granulocyte, lymphocyte, monocyte, neutrophil, T cell, B cell, natural killer cell, macrophage, cytokine, complement, acute phase protein, and immunoglobulin. Manual searches for additional studies were also conducted using the reference list(s) from germane articles identified in the computer searches.

All relevant articles that examined physical exercise training and immune system function in cancer survivors were considered. Studies were included in the review if they met the following three criteria. First, studies had to examine aerobic and/or resistance exercise training designed to improve cardiorespiratory fitness and/or muscular strength. Second, studies had to identify both the immune system component that was studied as well as the protocol used to assess it. Third, studies had to examine human subjects who performed physical exercise during or after cancer treatment.

Overall, six studies were found that met the inclusion criteria.36-41 These studies were published between 1994 and 2000 and all appeared as research articles. Table 4 summarizes information from the six studies including an extensive overview of the samples, designs, physical exercise interventions, assessments, and results separated by timing of the exercise intervention (i.e., during cancer treatment or posttreatment). An overview of these parameters, as well as the purposes and hypotheses of the studies, is presented below.

Table 4. Summary of Empirical Studies Examining the Effects of Physical Exercise Training on Immune System Function in Cancer Survivors
Authors Sample Design Physical exercise intervention Immune component(s) Immune assessment protocol(s) Results
Physical exercise during cancer treatment
 Dimeo et al.36 70 male and female autologous peripheral blood cell transplant patients with solid tumors (39 ± 10 y; 40 ± 11 y) Randomized controlled trial with usual care controls Supervised bed ergometer from first time of high dose chemotherapy until discharge (≈2 weeks), 7 ×/week @ 70% intensity, 30 min Peripheral blood neutrophils collected 12 h after the last exercise session Complete blood count ↓ duration of neutropenia and ↓ loss of physical performance after exercise intervention in exercise group compared to control group
 Shore et al.41 6 children with acute lymphoblastic leukemia or other type of neoplasm (14.0 ± 1 y) and 11 healthy controls (10.3 ± 2 y) Pretest-Posttest with matched controls (3 non-exercising cancer survivors and 11 healthy controls) Supervised cycling, soccer, skating, cross country skiing, swimming, or combination for 12 weeks, 3–4 ×/week @ 70–85% HRmax, 30 min Peripheral blood leukocytes, lymphocytes (CD3+, CD4+, CD8+, CD19+, CD25+, CD56+, CD122+) collected 36 h after the last exercise session Direct immunofluorescence, flow cytometry, 51Cr release assay, mitogen-stimulated lymphocyte proliferation Baseline: ↓ # leukocytes, CD3+, CD4+, CD8+, CD19+, CD25+, PHA-stimulated lymphocyte proliferation in exercise group cancer survivors receiving chemotherapy compared to healthy controls. Twelve weeks (not statistically significant): ↓ # CD3+, CD4+, CD8+, CD4+/CD8+, CD25+, ↑ IL-2 stimulated cytolytic activity in exercise group cancer survivors receiving chemotherapy compared to healthy controls; ↑ aerobic power after exercise intervention in exercise group cancer survivors receiving chemotherapy.
Physical exercise after cancer treatment
 Peters et al.38 24 female Stage I and II breast carcinoma survivors (49 ± 6.4 y) Pretest-posttest with no controls Supervised cycle ergometer for 5 weeks, 5 ×/week @ 60–86% HRmax, 30–40 min post-treatment. Then self-reported cycling for 6 months, 2–3 ×/week @ moderate intensity Peripheral blood lymphocytes (CD56+) Direct immunofluorescence, flow cytometry, 51Cr release assay ↔ # or % CD56+ at 5 or 29 weeks compared to baseline; ↑ NKCA (% lysis) at 29 weeks compared to baseline and 5 weeks.
 Peters et al.39 24 female Stage I and II breast carcinoma survivors (49 ± 6.4 y) Pretest-posttest with no controls Supervised cycle ergometer for 5 weeks, 5 ×/week @ 60–86% HRmax, 30–40 min post-treatment. Then self-reported cycling for 6 months, 2–3 ×/week @ moderate intensity Peripheral blood leukocytes, lymphocytes, monocytes, granulocytes Blood smear, functional test for the phagocytotic ability of monocytes ↔ # leukocytes, ↑ % granulocytes at 29 weeks compared to baseline and 5 weeks, ↓ %, # lymphocytes at 29 weeks compared to 5 weeks, ↓ % monocytes at 29 weeks compared to baseline and 5 weeks; ↑ phagocytosis (%, PI) of monocytes vs. RDSE at 29 weeks compared to baseline, ↔ phagocytosis (%, PI) vs. ADHE at any time point.
 Nieman et al.40 12 female breast carcinoma survivors (35–72 y) Randomized controlled trial with usual care controls Supervised walking for 8 weeks, 3 ×/week @ 75% HRmax, 60 min combined with supervised resistance training for 8 weeks, 2 sets of 12 repetitions of 7 exercises Peripheral blood leukocytes, lymphocytes (CD3+, CD3–CD16+CD56+), neutrophils Direct immunofluorescence, flow cytometry, 51Cr release assay ↔ # of leukocytes, CD3+, CD3–CD16+CD56+, neutrophils, NKCA and ↓ HR during fixed submaximal load, ↑ 16-min walk test distance, ↑ leg strength after exercise intervention in exercise group compared to control group.
 Na et al.37 35 stomach carcinoma patients (28–75 y) Randomized controlled trial with usual care controls Supervised range of motion exercise, pelvic tilting exercise, isometric-quadriceps-setting exercise 3 ×/day, 30 min. Then supervised arm or cycle ergometer for 2 weeks, 10 ×/week, 30 min Peripheral blood lymphocytes 51Cr release assay ↑ NKCA (% lysis) after exercise intervention in exercise group compared to control group
  • y: years; min: minutes; h: hours; HRmax: maximum heart rate; NKCA: natural killer cell cytolytic activity; PI: phagocytosis index; RDSE: receptor destroying enzyme-treated sheep erythrocytes; ADHE: anti D-loaded human erythrocytes; ↑: increase; ↓: decrease; ↔: no change.
  • a Nonsignificant.

Purposes and hypotheses

The purposes of the six studies were similar. All six studies examined the effect of physical exercise training on immune system function in cancer survivors during or after cancer treatment.36-41 For example, two studies examined the effect of physical exercise training on NK cell cytolytic activity in breast carcinoma survivors.38, 40 Similarly, the hypotheses tested in the six studies were comparable. Specifically, the investigators hypothesized that moderate intensity physical exercise training would improve immune system function in cancer survivors during or after cancer treatment.36-41

Samples and designs

Participants in the six studies were varied. Two out of six studies examined cancer survivors during treatment,36, 41 whereas four studies examined cancer survivors posttreatment.37-40 Two studies examined the same sample of breast carcinoma survivors who had Stage I or II disease,38, 39 while another study examined a different sample of breast carcinoma survivors.40 The other studies investigated autologous peripheral blood cell transplant survivors with solid tumors,36 stomach carcinoma survivors,37 and children with a history of acute lymphoblastic leukemia, Ewing Sarcoma, or non-Hodgkin lymphoma.41 Three out of six studies examined females,38-40 one study investigated both males and females36 and two studies did not report the gender of their subjects.37, 41 The sample sizes for the studies ranged from 641 to 70 survivors36 with a mean of 28 survivors. Only one study performed a statistical power calculation to determine the number of subjects required in each group to detect a clinically significant difference in the primary endpoint.36

Study designs in the six studies also varied. Three studies were randomized controlled trials (RCTs) with usual care controls,36, 37, 40 two studies were a pretest-posttest design with no controls,38, 39 and one study was a pretest-posttest design with matched controls (i.e., non-exercising cancer survivors and exercising healthy persons).41

Physical exercise interventions

Several physical exercise interventions were utilized. All six studies initiated exercise that was consistent with current guidelines recommended for the development of physical fitness in cancer survivors.42 The frequency of the exercise was five times per week for two studies38, 39 and three or four times per week for an additional two studies.40, 41 Other studies reported an exercise training frequency of seven times per week36 and 10 times per week.37 The intensity of the exercise was between 60% and 80% of each subjects' maximum heart rate in five out of six studies.36, 38-41 One study failed to provide exercise training intensity.37 The time spent on exercise during each training session was between 30 and 40 minutes in five studies36-39, 41 and 60 minutes in one study.40 The length of the exercise interventions were 29 weeks,38, 39 12 weeks,41 8 weeks,40 or 2 weeks.36, 37 The primary exercise mode in three out of six studies was cycle ergometer.36, 38, 39 One study combined walking and resistance training exercise,40 while one study combined range of motion exercises, pelvic tilting exercises, isometric quadriceps-setting exercises, and cycle ergometer.37 The other study provided the subjects with a choice of several aerobic activities.41 The exercise was supervised for the entire duration in four studies,36, 37, 40, 41 whereas two studies provided supervision for the first five weeks of the intervention.38, 39

Assessments

Several physical fitness assessments were used to document an exercise training effect. Three out of six studies used a graded or progressive bicycle ergometer test until exhaustion,38, 39, 41 whereas one study used a graded treadmill stress-test until exhaustion.36 One study used a combination of a symptom-limited treadmill exercise test, 16-minute walk test, and leg extension strength test.40 The physical fitness assessment method was not reported in one study.37

A variety of cellular immune system components were assessed. All six studies obtained immune system cells from samples of peripheral blood.36-41 One study assessed NK cell (CD56+) counts and cytolytic activity,38 while one study each measured NK cell cytolytic activity37 and neutrophils.36 One study assessed all leukocytes including lymphocyte, monocyte, and granulocyte populations,39 whereas one study measured all leukocytes including the number of NK cells (CD56+), T cells (CD3+, CD4+, CD8+, CD25+, CD122+) and B cells (CD19+).41 Finally, one study assessed all leukocytes including the number of T cells (CD3+), NK cells (CD3–CD16+CD56+), and granulocytes.40

Several immunologic protocols were used to assess immune system function. Three studies employed a combination of direct immunofluorescence and flow cytometry.38, 40, 41 One study used a combination of a blood smear and functional test for the phagocytotic ability of monocytes,39 while one study used complete blood counts.36 Four studies37, 38, 40, 41 utilized the 51Cr release assay and one study41 used PHA- and PWM-induced lymphocyte proliferation techniques.

RESULTS

In general, the studies that examined the effect of physical exercise training on immune system function in cancer survivors reported favorable outcomes. The subjects assigned to the physical exercise training group had lower heart rates during a fixed submaximal load, increased 16-minute walk distances, and increased leg strength compared to the control group after the exercise intervention.40 In addition, trained subjects had a decreased loss of physical performance36 and increased aerobic power41 compared to untrained subjects. Three studies did not report changes in cardiorespiratory fitness and/or muscular strength.37-39

Similarly, the studies reported favorable immune system outcomes. More specifically, four out of six studies reported statistically significant improvements in immune system function as a result of exercise. The immunologic benefits that have been shown include improvements in NK cell cytolytic activity,37, 38 monocyte function,39 proportion of circulating granulocytes,39 and duration of neutropenia.36 Notably, these statistically significant results occurred in the studies despite the fact that they had an average of only 28 participants each. In contrast, two studies found no statistically significant improvements in immune function as a result of exercise. In fact, Nieman et al. found no statistically significant change in NK cell cytolytic activity or the proportion of T and NK cells,40 while Shore and Shephard reported non-significant decreases in T cell populations as a result of exercise.41

Limitations of Past Research and Directions for Future Research

Although the extant literature suggests that physical exercise training may have a positive influence on several immune system components important in cancer defense, there are several limitations that should be considered when interpreting the results and planning future research. Moreover, there are many unexplored issues due to the nascency of this field that warrant further investigation. Some limitations of past research and directions for future studies are presented below.

Samples

Four important sample limitations were identified. First, all six studies used convenience samples.36-41 This method of sampling is problematic in that there is no defined population from which the sample is drawn. As a result, the generalizability of the findings is limited because it is difficult to estimate the amount and nature of the selection bias that may exist in the sample. Second, five out of six studies collected data using small numbers of subjects.37-41 For example, in those five studies, an average of 20 subjects were assessed, of which 10 were included within the experimental condition. Small numbers of subjects reduce the power of a study and preclude the use of multivariate statistical techniques.43 In addition, only one study performed a statistical power calculation to determine the number of subjects required to detect a clinically significant difference between groups in the primary endpoint.36 Consequently, future exercise immunology research should attempt to recruit larger, random samples from a defined population of cancer survivors. Researchers should also give sufficient attention to statistical power in the planning stages of research to ensure that they will be able to detect a significant difference between groups in the primary endpoints, if one does indeed exist.

Third, subjects within all six studies were heterogeneous with respect to one or more characteristics including age, gender, and cancer site. Importantly, research has shown that several components of the immune system are significantly influenced by such variables.44, 45 For instance, increasing age has been associated with decreases in the total number of peripheral blood lymphocytes and T cells (CD4+, CD8+), T cell (CD45RA+) function, T cell responses to mitogenic stimulation, IL-2 production by T cells, and NK cell cytolytic activity, to name a few.46 In addition, four studies did not report information about the stage of cancer,36, 37, 40, 41 while five studies did not provide data regarding subject treatment protocols.37-41 Therefore, future exercise immunology research should attempt to recruit homogenous samples of cancer survivors to reduce misinterpretations of any exercise-induced immune system response. Researchers must also describe, in sufficient detail, information regarding the phase of treatment, as was done to a limited extent in one study.36 In particular, researchers should attempt to identify cancers using the TNM classification system, which assesses tumors in three ways: extent of the primary tumor, absence or presence of regional lymph node involvement, and absence or presence of distant metastases.1 In addition, investigators should delineate specific details regarding the treatment protocols (e.g., mastectomy and axillary node dissection, cytotoxic chemotherapy, and hormonal therapy) and the timing of these interventions in relation to exercise (i.e., exercise during cancer treatment versus posttreatment). Due to the profound and prolonged immune effects of allogenic red blood cell transfusions47 commonly required in the perioperative period or during chemotherapy administration, transfusion data must also be obtained. Such information will allow exercise immunologists to accurately compare exercise-induced immune system alterations and, ultimately, help to reduce misinterpretation of the immune response in cancer survivors.

Fourth, three out of six studies have examined breast carcinoma survivors,38-40 while one study each assessed mixed cancer survivors recovering from autologous peripheral blood stem transplant,36 children with mixed cancers41, and stomach carcinoma survivors.37 The unique demographics, pathology, surgical procedures, and treatment protocols make it unwise to generalize the results from one cancer site to another. Thus, future exercise immunology research should examine common cancers such as prostate, lung, colon, kidney, bladder, and uterine carcinomas and, in particular, malignancies that have been shown to have a strong association with the immune system (e.g., melanoma).

Design

One important design limitation was identified. Two out of six studies utilized a pretest-posttest design with no controls38, 39 and one study used a pretest-posttest design with matched controls.41 These designs are problematic in that they leave findings potentially vulnerable to numerous threats to internal validity (e.g., circadian and seasonal changes in immune system function).48 Therefore, future exercise immunology research should attempt to use RCTs with usual care controls or wait-list controls, which will help to reduce the influence of confounding variables.43 RCTs have also been proposed as the strongest design by which to examine biomarkers in exercise and cancer research.49

Physical Exercise Interventions

Two main limitations of the physical exercise interventions were identified. First, three out of six studies prescribed an intervention that lasted for a period of less than 12 weeks.36, 37, 40 This intervention period is relatively short, particularly if the intent is to examine the effect of physical exercise on immune system function, and subsequent risk of cancer recurrence and/or secondary malignancies and increase in survival times. Thus, future exercise immunology research should attempt to assess interventions that last for a minimum period of 12 weeks. Second, three out of six studies employed partially unsupervised training interventions.38, 39, 41 Self-report measures of physical exercise may lead to inaccurate data regarding the frequency, intensity, and duration of exercise that was performed.50 Therefore, future exercise immunology research should attempt to employ supervised exercise interventions with close monitoring of exercise parameters.

Assessments

Two limitations of the physical fitness assessments were identified. One study did not provide any information about the physical fitness assessment(s).37 Clearly, it is imperative that exercise immunology studies in cancer survivors provide such information. Second, two studies performed a preintervention assessment but no postintervention assessment.38, 39 Accordingly, future exercise immunology research should attempt to assess cardiorespiratory fitness and/or muscular strength before and after the exercise intervention to document changes in physical fitness.

Seven main limitations of the immune system components and methods were identified. First, no study has related exercise-induced alterations in immune system function to important clinical outcomes for cancer survivors. Examples of germane clinical outcomes include the toxicity of anticancer treatments, the risk of recurrence and/or death, and the onset of late effects, to name only a few. Accordingly, future exercise immunology research in cancer survivors should attempt to correlate changes in immune system function to meaningful clinical outcomes.

Second, immune cells from all six studies were characterised using samples of peripheral blood. Peripheral blood samples may not be representative of the condition of the whole body since a large percentage of all leukocytes are normally found outside of the circulating peripheral blood.51 As a result, exercise immunologists should be cautious in the interpretation of such data if they are used to predict immune system component alterations in the lymphoid organs.51 In addition, future exercise immunology research should attempt to establish whether exercise-induced alterations in immune system components from peripheral blood samples are similar to those obtained from the lymphoid tissues and/or other body fluids. Notably, animal models may be helpful in this regard.52, 53

Third, blood collection protocols that were used across the six studies were partly inconsistent. For example, five out of six studies did not provide dietary intake information on the days prior to blood collection.36, 38-41 Moreover, no study provided information pertaining to medications and/or sleeping patterns prior to blood collection. Because cellular and humoral immune system components are dramatically influenced by diet and nutritional status,54, 55 medication,45 sleep patterns,45 and smoking status,45 subject behavior prior to blood collection is an obvious source of variability. Consequently, future exercise immunology research should attempt to establish standards for subject behavior in the periods several days before blood collection and during the time-course of the sampling procedure itself.56 Such protocols might ask participants to consume the same diet on the three days prior to each blood collection session throughout the study.

Fourth, three out of six studies assessed immune system components before and after the intervention,36, 40, 41 whereas three studies measured immune system function before, at some point during, and after the intervention.37-39 As a result, there is a paucity of information regarding alterations in immune system components during the exercise intervention. Future exercise immunology research should attempt to assess immune system function at several time points throughout the intervention (i.e., multiple time point analysis). Such a regimen might include assessments prior to program initiation, at program midpoint, at program endpoint, and postprogram follow-up.

Fifth, four out of six studies did not report the time at which blood was drawn in relation to exercise.37-40 One study each collected blood specimens 12 hours36 and 36 hours41 after the final exercise session. Given that the effect of exercise on the immune system both during and after an exercise bout is still unknown, future exercise immunology research should report the time at which blood specimens are collected. Moreover, studies should attempt to collect blood specimens and thus measure immune system function at several time points during and following completion of exercise.

Sixth, several cancer-related immune system components have yet to be examined. More specifically, phenotype analysis to determine the absolute numbers and/or percentages of leukocyte sub-populations and lymphocyte subsets were relatively limited, and the influence of exercise on several immune cell types remains unknown. In addition, only one study examined lymphocyte proliferation in response to antigenic stimulation,41 and no study assessed the oxidative capacity of neutrophils or B cell function. Similarly, no study assessed acute phase proteins (i.e., C-reactive protein) or cytokines (IL-1, IL-6, IFN-α, IFN-γ, and TNF-α), all of which may be important in anticancer defense6 and several of which are influenced by exercise.57 Therefore, future exercise immunology research should attempt to assess a wide variety of cancer-related immune system components. Phenotype analysis, functional assays, and cytokine analysis should be incorporated into future studies.

Seventh, the presentation of immune system results was partly inconsistent with current guidelines in exercise immunology.56 For instance, NK cell cytolytic activity assay results can be presented as single or multiple effector cell to target cell (E:T) ratios). Two out of six studies presented results using one E:T ratio,37, 38 while one study each presented results using two E:T ratios40 and three E:T ratios.41 Multiple ratios allow the researcher to present results as lytic units. and single ratios can be influenced by the position of the titration curve.58 Therefore, future exercise immunology research should attempt to present NK cell assay results as a minimum of two, but preferably three or more, E:T ratios. Another drawback in the presentation of immune function results involves cellular subset analyses. Results from lymphocyte subset analyses can be presented as absolute numbers and/or percentages of cells. Importantly, Smith has recommended that such results be presented as both absolute numbers and percentages to allow researchers to make corrections in retrospective analysis (e.g., meta-analysis).56 Unfortunately, two studies reviewed herein presented the absolute number but not the percentage of lymphocytes in each subset.40, 41

Other recommendations for future research

Research suggests that the effects of physical exercise on immune system function may be mediated by a number of potential mechanisms. These include neuroendocrine hormones,59-61 elevated body temperature,62 energy sources,63 and autocrine or paracrine molecules.64 Recently, neuroendocrine components have received the most research attention in the exercise immunology literature. Foremost among these components are catecholamines, growth hormone, cortisol, β-endorphin, and sex steroids.61-65 Given the diverse interactions between these biologic systems and their potential role in cancer defense, exercise immunologists should be aware of and consider the impact of these complex interactions in future exercise immunology trials in cancer survivors. Knowledge of fundamental immunology, endocrinology, cell biology, and nutrition may help exercise immunologists to more fully appreciate the role of physical exercise in reducing the risk of cancer recurrence and/or secondary malignancies and increasing survival times.

As one example, cytotoxic T lymphocytes (CTL) have been shown to mediate tumor cell apoptosis via granule-mediated exocytosis.66 According to this mechanism, an interaction between granzyme B (GrB) and the mannose 6-phosphate/insulin-like growth factor receptor (CI-MPR) is essential for cell surface binding, uptake, and induction of tumor cell apoptosis.67 Interestingly, insulin-like growth factors (IGF) have been shown to stimulate the release of CI-MPR from breast carcinoma cell lines and metastatic breast carcinoma cells.68 An IGF-induced reduction in the number of CI-MPRs on breast carcinoma cells may mean that there are fewer GrB – CI-MPR interactions, reducing the susceptibility of breast carcinoma cells to CTL-induced apoptosis. Importantly, physical exercise has been shown to reduce IGF-1. For example, Schmitz et al. reported that a 15-week resistance exercise program resulted in decreased IGF-1 in female subjects despite increased lean mass.69 By inference, exercise-induced decreases in IGF-1 may decrease the release of CI-MPR from breast carcinoma cells, increase the number of GrB – CI-MPR interactions between CTL and breast carcinoma cells, and promote CTL-induced cancer cell apoptosis. Mechanistic studies of exercise-induced cancer control are needed.

CONCLUSIONS

A strong rationale exists for examining physical exercise and cancer recurrence/survival through effecting immune system function in cancer survivors. To our knowledge, six studies have examined this issue to date. The results of these studies suggest that physical exercise training may improve a number of immune system parameters that may be important in cancer defense. However, these studies have limitations in sample size, study design, physical exercise interventions, physical fitness assessments, and immunologic assessments. Moreover, many unanswered questions remain. Additional research is needed to determine if exercise in cancer survivors may reduce the risk of recurrence and/or secondary malignancies and increase survival times.