Effects of high‑intensity interval training on physical function and quality of life in breast cancer patients: a systematic review and meta-analysis

Abstract

Background

Breast cancer remains the most common malignancy among women worldwide. Despite advances in screening and treatment that have improved five-year survival, patients often experience weight gain, reduced cardiorespiratory fitness, metabolic disturbances, and impaired quality of life during or after therapy. Moderate‑intensity continuous training (MICT) is effective but time‑consuming, while high‑intensity interval training (HIIT) offers a time‑efficient alternative that may provide greater cardiopulmonary stimulation. However, evidence on its effects on body composition, quality of life, and inflammatory markers in breast cancer patients remains inconclusive.

Methods

A comprehensive search of PubMed, Embase, Web of Science, the Cochrane Library, CNKI, Wanfang and VIP databases was conducted up to July 31, 2025 to identify randomized controlled trials (RCTs) comparing HIIT with MICT or usual care (UC) in women aged ≥ 18 years diagnosed with breast cancer. Eligible interventions involved HIIT, while control groups received MICT or UC. The primary outcome was peak oxygen uptake (VO₂peak), reflecting cardiorespiratory fitness, and quality of life, whereas secondary outcomes included body composition, and the inflammatory marker interleukin‑6(IL‑6). Standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated, and fixed or random‑effects models were applied according to heterogeneity levels.

Results

A total of ten randomized controlled trials (providing 13 comparison arms) involving 330 participants were analyzed. Compared with control interventions, HIIT significantly increased VO₂peak (SMD = 0.74; 95% CI [0.47,1.01]; I²=9.2%), with further subgroup analysis showing a significant improvement versus UC (SMD = 0.85; 95% CI [0.54,1.15]) but no significant difference compared with MICT (SMD = 0.38; 95% CI [–0.18,0.94]). No significant changes were found in BMI (SMD=–0.16; 95% CI [–0.45,0.13]), fat mass (SMD=–0.23; 95% CI [–0.51,0.06]), lean body mass (SMD = 0.22; 95% CI [–0.06,0.51]) or body fat percentage (SMD=–0.32; 95% CI [–0.88,0.24]). Regarding quality of life, functional well‑being (SMD = 0.89; 95% CI [0.39,1.39]) and social well‑being (SMD = 0.79; 95% CI [0.29,1.28]) improved significantly, whereas emotional well‑being improved only when HIIT was compared with UC (SMD = 0.65; 95% CI [0.08,1.23]). Physical well‑being did not differ significantly (SMD = 0.24; 95% CI [–0.24,0.71]). HIIT did not significantly reduce IL‑6 levels (SMD=–0.22; 95% CI [–0.65,0.22]).

Conclusions

HIIT appears to be safe and significantly improved cardiorespiratory fitness (VO₂peak) compared with usual care but showed no significant effects on body composition or IL-6 levels. Benefits in some quality-of-life domains were observed but based on limited evidence. HIIT did not outperform moderate-intensity continuous training.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13102-026-01811-9.

Introduction

Breast cancer is one of the most commonly diagnosed malignancies among women and a leading cause of cancer‑related mortality worldwide [1]. According to the International Agency for Research on Cancer (IARC), approximately 2.3 million new breast cancer cases and about 685,000 deaths occurred globally in 2020, accounting for approximately 16% of all female cancer deaths. By 2040, the global burden is projected to exceed 3 million new cases and 1 million deaths annually [2]. Advances in early detection and comprehensive treatment have markedly improved five‑year survival rates, reaching up to 90% in many countries, alongside a sustained decline in breast cancer–specific mortality over the past two decades [3]. However, these therapies are often associated with long‑term side effects. Chemotherapy, particularly taxanes and anthracyclines, may cause peripheral neuropathy, cognitive impairment and cardiotoxicity. Targeted therapies may lead to left‑ventricular dysfunction, while endocrine therapies commonly result in osteoporosis, menopausal symptoms, genitourinary syndrome and metabolic disturbances. Furthermore, these therapies can adversely affect mental health, body weight and overall quality of life [4]. Therefore, interventions that preserve physical function and quality of life while reducing the risks of recurrence and comorbidities are critically needed.

Physical activity is a safe and effective non‑pharmacological intervention that improves cardiorespiratory fitness, muscle strength, posture, psychological well‑being, and social engagement, while also reducing the risks of cardiovascular events and cancer recurrence [5].International Multidisciplinary Roundtable recommended that cancer survivors engage in at least three 30‑minute sessions of moderate-intensity aerobic exercise per week, combined with 8–12-weeks programs of aerobic and resistance training, to counteract many treatment‑related adverse effects. ASCO guidelines emphasize that patients undergoing curative-intent cancer treatment should avoid physical inactivity and, when feasible, participate in regular aerobic exercise combined with resistance training, with programs individualized according to treatment status and symptom burden [5, 6]. Moreover, a recent systematic review and meta‑analysis in patients with non‑small cell lung cancer reported that exercise interventions generally improve functional capacity, pulmonary function, and certain aspects of quality of life, although the strength of evidence is not consistent across cancer types and exercise modalities [7].For patients undergoing chemotherapy or endocrine therapy, however, moderate-intensity exercise can be time‑consuming and may provide limited cardiopulmonary stimulation, making adherence difficult. Thus, there is a need for interventions that efficiently improve cardiorespiratory fitness, reduce fatigue, and enhance quality of life within a limited time.

High‑intensity interval training (HIIT) involves short bouts of high-intensity exercise lasting 15–120 s near maximal oxygen uptake or heart rate, interspersed with equal or longer periods of low-intensity recovery, generally reaching 85–100% of VO₂max. Compared with moderate‑intensity continuous training (MICT), HIIT is more time efficient and provides stronger cardiopulmonary stimulation and higher metabolic load [8]. In populations with cardiovascular disease, obesity, diabetes, and cancer, HIIT has been shown to markedly increase VO₂peak, improve insulin sensitivity, and enhance metabolic health [9–12]. Beyond these cardiometabolic benefits, emerging evidence suggests that HIIT may also exert anti-inflammatory effects relevant to cancer prognosis. Inflammatory pathways play a key role in breast cancer symptoms and prognosis, with chronic low‑grade inflammation, particularly elevated interleukin‑6 (IL‑6), linked to tumor progression [13]. Exercise-induced reductions in inflammatory cytokines have been associated with improvements in fatigue and other symptoms [14]. Clinical studies suggest that HIIT may modulate the tumor microenvironment by regulating inflammatory IL‑6, TNF‑α and IL‑10, which could contribute to improve prognosis [15].

RCTs indicate that HIIT can achieve greater short‑term improvements in VO₂peak and help prevent declines in physical capacity compared with MICT [16–22]. Evidence generally supports beneficial effects of HIIT on quality of life in breast cancer patients, although results are heterogeneous due to differences in cancer stage, treatment regimen, age, and outcome measures [17, 23–25].Findings on body composition and IL-6 are also inconsistent, with some studies reporting improvements and others showing no significant change [14, 15, 17].

Given these uncertainties, the study systematically reviewed and meta‑analyzed RCTs of HIIT in breast cancer patients. It evaluated its effects on cardiorespiratory fitness, body composition, quality of life, and the inflammatory marker IL‑6, compared with UC and with MICT, to provide evidence‑based guidance for exercise rehabilitation.

Methods

Eligibility criteria

Studies were eligible if they met the following criteria: 1.Study design: RCT. 2.Participants: women aged ≥ 18 years with pathologically confirmed breast cancer, regardless of disease stage, surgical history, or treatment modality. 3.Intervention: HIIT as the sole intervention in the experimental group. 4.Control: UC or non‑HIIT exercise (e.g. MICT)0.5.Outcomes: at least one outcome among VO₂peak, quality of life, body composition or inflammatory markers.

Exclusion criteria were: 1.Animal or non‑interventional studies. 2.Reviews, conference abstracts, or reports lacking complete data. 3.Insufficient or unavailable data for analysis. 4.Interventions combining HIIT with other treatments. 5.Duplicate publications or inaccessible full texts.

Information sources and search strategy

We searched PubMed, Embase, Web of Science, the Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang and VIP databases for RCTs up to July 31, 2025. The search strategies combined subject terms and free‑text terms customized for each database. Terms included “high‑intensity interval training,” “breast cancer,” “randomized controlled trial” and their synonyms. No restrictions were imposed on publication date or language. The detailed strategies are provided in Supplementary Table S1. This review followed the PRISMA guidelines and was registered in PROSPERO (CRD420251159884).

Selection process and data collection process

Two reviewers independently screened titles and abstracts to identify potentially relevant studies, followed by full-text assessment according to the predefined criteria. Any disagreements were resolved through discussion or consultation with a third reviewer. Extracted data included study design, sample size, control intervention, training protocol and outcome measures. For each continuous outcome, post-intervention endpoint values were extracted. VO₂peak and quality-of-life outcomes were defined as the primary outcomes, while body composition indices (BMI, fat mass, lean body mass, and body fat percentage) and the interleukin-6 (IL-6) were considered secondary outcomes. Discrepancies in extracted data were resolved by consensus with a third reviewer. Studies with essential missing data were excluded from the meta‑analysis.

Risk of bias assessment

The same two reviewers independently assessed the risk of bias of the included randomized controlled trials using the revised Cochrane Risk of Bias tool for randomized trials (RoB 2). Bias was evaluated at the outcome level across the following five domains: bias arising from the randomization process; bias due to deviations from intended interventions; bias due to missing outcome data; bias in measurement of the outcome; and bias in selection of the reported result.

Each domain was judged as “low risk,” “some concerns,” or “high risk”, and an overall risk-of-bias judgment was assigned for each study according to the RoB 2 algorithm. Any disagreements between the reviewers were resolved through discussion, with consultation of a third reviewer when necessary.

Statistical analysis

Meta‑analyses were conducted using R (version 4.3.3) with R Studio and Review Manager (version 5.3). For continuous outcomes, standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated. Heterogeneity was assessed using the I² statistic. A fixed‑effect model was applied when heterogeneity was not significant (P > 0.05, I² < 50%); otherwise, a random‑effects model was used. A two‑tailed P < 0.05 was considered statistically significant.

Results

Literature search results and study characteristics

A total of 433 records were retrieved from the database search. After applying the inclusion and exclusion criteria, 10 RCTs were included in the final analysis (Fig. 1) [16–23, 26, 27]. The key characteristics of the included studies are summarized in Table 1. Participants were predominantly middle‑aged women (mean age 44–67 years). Baseline characteristics, including age, weight and BMI, were generally comparable between intervention and control groups. Sample sizes were relatively small. Most patients had stage I to III breast cancer, with stages I and II being most common. All studies employed randomized controlled design. Three of them (Coletta 2021, Isanejad 2023, Northey 2019) each included two control arms (MICT and UC), contributing a total of 13 comparison arms for meta-analysis. HIIT programs typically consisted of two to three sessions per week for 8–12 weeks, while controls received UC or MICT.Detailed information on the intervention protocols and the specific methods used to measure VO₂peak is provided in Tables S2 and S3.

Fig. 1.

Study inclusion flow chart. Flow chart illustrating the study identification, screening, eligibility assessment, and inclusion process in accordance with the PRISMA 2020 guidelines

Table 1.

General baseline data

Author, year	Sample Size (I/C)	Age (years, I/C)	Height (cm, I/C)	Weight (kg, I/C)	BMI (kg/m², I/C)	Cancer Stage (n, I/C)	Control Measure
Bell 2021 [16]	10/10	49 ± 4/51 ± 5	165 ± 5/166 ± 7	66 ± 16/74 ± 14	25.1 ± 5.6/27.1 ± 6.8	I 5/1 II 5/9	MICT
Coletta 2021 [26]	12/10	64.4 ± 6.2/66.1 ± 13.6	N/R	N/R	32.3 ± 8.6/32.0 ± 5.6	N/R	MICT
Coletta 2021 [26]	12/11	64.4 ± 6.2/62.6 ± 7.0	N/R	N/R	32.3 ± 8.6/29.9 ± 2.5	N/R	UC
Gonzalo-Encabo 2022 [27]	15/15	49.1 ± 7.9/44.7 ± 11.2	156.5 ± 6.6/160.3 ± 9.9	80.9 ± 17.7/74.5 ± 18.8	33.1 ± 7.6/30.1 ± 7.7	I 1/1 II 5/4 III 9/10	UC
Isanejad 2023 [17]	10/10	44.00 ± 9.14/44.90 ± 5.02	N/R	68.08 ± 8.87/68.94 ± 11.18	26.50 ± 3.48/27.37 ± 4.88	I 1/0 II 4/5 III 5/5	UC
Isanejad 2023 [17]	10/10	44.00 ± 9.14/46.29 ± 6.29	N/R	68.08 ± 8.87/72.85 ± 11.07	26.50 ± 3.48/28.10 ± 5.14	I 1/1 II 4/4 III 5/5	MICT
Klavina 2025 [18]	13/11	47.46 ± 8.41/52.45 ± 9.73	N/R	N/R	N/R	II 8/8 III 3/3	UC
Lee 2019 [19]	15/15	49.1 ± 7.9/44.7 ± 11.2	156.5 ± 6.6/160.3 ± 9.9	80.9 ± 17.7/74.5 ± 18.8	33.1 ± 7.6/30.1 ± 7.7	I 1/1 II 5/4 III 9/10	UC
Lee 2021 [23]	15/15	49.1 ± 7.9/44.7 ± 11.2	156.5 ± 6.6/160.3 ± 9.9	80.9 ± 17.7/74.5 ± 18.8	33.1 ± 7.6/30.1 ± 7.7	I 1/1 II 5/4 III 9/10	UC
Northey 2019 [20]	6/6	60.3 ± 8.1/61.5 ± 7.8	165.6 ± 5.8/163.5 ± 5.5	69.5 ± 16.1/75.6 ± 7.7	N/R	I2/3 II 2/3 III 2/0	UC
Northey 2019 [20]	6/5	60.3 ± 8.1/67.8 ± 7.0	165.6 ± 5.8/165.6 ± 5.6	69.5 ± 16.1/68.8 ± 11.5	N/R	I 2/1 II 2/4 III 2/0	MICT
Ochi 2022 [21]	25/25	48 ± 6/49 ± 5	N/R	N/R	21.0 ± 2.2/20.9 ± 2.0	I 18/18 II 7/7	UC
Samhan 2021 [22]	30/30	49.7 ± 8.9/48.9 ± 7.7	N/R	92.23 ± 8.34/91.56 ± 8.72	33.22 ± 1.29/33.48 ± 1.45	I 18/15 II 7/11 III 5/4	UC

Three studies (Coletta 2021, Isanejad 2023, Northey 2019) provided two comparison arms each; the remaining seven provided one each. Total:10 studies,13 comparison arms.Values are presented as mean ± standard deviation

I Intervention group, C Control group, BMI Body mass index, N/R Not reported, RCT Randomized controlled trial, MICT Moderate-intensity continuous training, UC Usual care

Risk of bias assessment of included studies

All included trials were randomized controlled studies. Using the RoB 2 tool, most studies were judged as having low risk of bias or some concerns across domains. Higher concerns were mainly observed in the domains of the randomization process and deviations from intended interventions, while missing outcome data and measurement of outcomes were generally rated as low risk. As the interventions involved exercise training, blinding of participants and personnel was often not feasible, which may have contributed to higher risk judgments in the deviations from intended interventions domain (Figs. 2 and 3).Overall, the included studies were of acceptable methodological quality.

Fig. 2.

Risk of bias assessment. Risk of bias was assessed using the RoB 2. The traffic-light plot displays domain-level and overall risk-of-bias judgments for each included study across the five RoB 2 domains

Fig. 3.

Risk of bias graph. The bar chart summarizes the proportion of studies rated as low risk, some concerns, or high risk for each RoB 2 domain and for overall bias

VO₂peak

VO₂peak is a key indicator of cardiorespiratory fitness and exercise tolerance. Nine comparison arms from seven RCTs evaluated the effect of HIIT on VO₂peak(n = 231) [16–19]. Pooled analysis showed that HIIT significantly improved VO₂peak compared with control interventions (SMD = 0.74; 95% CI [0.47,1.01]), with low heterogeneity (I²=9.2%). A fixed‑effect model was therefore used. Subgroup analysis revealed a significant improvement compared with UC (SMD = 0.85; 95% CI [0.54,1.15]), whereas the difference versus MICT was not statistically significant (SMD = 0.38; 95% CI [–0.18,0.94]). Sensitivity analyses confirmed the robustness of the result, suggesting that HIIT may enhance cardiorespiratory fitness in breast cancer patients(Figs. 4, 5 and 6).

Fig. 4.

Meta-analysis of VO₂peak. Forest plot showing standardized mean differences (SMDs) with 95% confidence intervals (CIs) for the effect of high-intensity interval training (HIIT) on VO₂peak compared with control interventions

Fig. 5.

Forest plot of pooled subgroup analysis of VO₂peak. Forest plot of subgroup analyses comparing the effects of high-intensity interval training (HIIT) versus usual care and versus moderate-intensity continuous training on VO₂peak

Fig. 6.

Sensitive analysis of VO2peak. Sensitivity analysis conducted by sequentially excluding individual studies to assess the robustness of the pooled effect estimate for VO₂peak

Quality of life

Quality of life was evaluated using the Functional Assessment of Cancer Therapy – General (FACT‑G) scale, which includes four dimensions: emotional, functional, physical and social well-being. Higher scores indicate better quality of life.

Emotional well-being constitutes a critical dimension in assessing a patient’s emotional state and psychological adaptability.Two trials assessed the emotional well‑being domain(n = 60) [17, 23]. Meta-analysis indicated moderate heterogeneity (I²=40.2%) and no significant overall improvement (SMD = 0.44; 95% CI [–0.04,0.92]). However, subgroup analysis revealed a significant benefit of HIIT compared with UC (SMD = 0.65; 95% CI [0.08,1.23]). No significant differences were observed between subgroups. Sensitivity analysis excluding the MICT control study (Isanejad 2023) yielded consistent results (SMD = 0.65; 95% CI [0.08,1.23]; P = 0.0263), supporting the reliability of the finding(Fig. 7,a).

Fig. 7.

Forest plot of pooled subgroup analysis of quality of life. Forest plots of pooled standardized mean differences (SMDs) with 95% confidence intervals for quality-of-life domains, including (a) emotional well-being, (b) functional well-being, (c) physical well-being, and (d) social well-being

Functional well‑being reflects the ability to perform daily activities and maintain independence. Two trials evaluated this dimension(n = 60) [17, 23]. The pooled analysis showed low heterogeneity (I²=3.8%) and demonstrated a significant improvement following HIIT (SMD = 0.89; 95% CI [0.39,1.39]). Subgroup analysis comparing HIIT with UC confirmed this effect (SMD = 1.06; 95% CI [0.46,1.67]). Differences between subgroups were not significant. Sensitivity analyses suggested that individual studies might partially influence the pooled results but did not alter the overall conclusion(Fig. 7,b).

Physical well‑being reflects perceptions of pain, fatigue, and other physical symptoms. Two trials were included in this analysis(n = 60) [17, 23]. Meta‑analysis indicated no significant effect of HIIT (SMD = 0.24; 95% CI [–0.24,0.71]) with low heterogeneity (I²=4.4%). Subgroup analysis comparing HIIT with UC also showed no significant difference (SMD = 0.45; 95% CI [–0.11,1.02]). Sensitivity analyses confirmed the robustness of these results(Fig. 7,c).

Social well‑being represents satisfaction with interpersonal relationships and family or social support. Two studies reported this outcome(n = 60) [17, 23]. Meta-analysis indicated a significant improvement in social well-being following HIIT (SMD = 0.79; 95% CI [0.29,1.28]) with low heterogeneity. Subgroup analysis comparing HIIT with UC confirmed this effect (SMD = 0.81; 95% CI [0.23,1.39]), no significant subgroup differences were observed. Sensitivity analyses affirmed the stability of these findings(Fig. 7,d).

Body composition

Six trials reported body composition outcomes, including BMI, fat mass, lean body mass and body fat percentage [16–18, 21, 22, 27]. Body composition outcomes were assessed using different measurement methods across the included studies. Details of the assessment techniques used in each study are summarized in Table 2.

Table 2.

Body composition assessment methods

Study (Author, year)	Body composition variables included in this review	Assessment method
Bell 2021 [16]	BMI, fat mass, lean body mass	DXA(Lunar Corporation, Madison WI)
Gonzalo-Encabo 2022 [27]	fat mass, lean body mass, body fat percentage	BIA(InBody 770,Cerritos, CA)
Isanejad 2023 [17]	BMI, fat mass, lean body mass, body fat percentage	Skinfold measurements
Klavina 2025 [18]	BMI	N/R
Ochi 2022 [21]	BMI, fat mass, lean body mass, body fat percentage	BIA(MC-780 A-N; Tanita, Tokyo, Japan)
Samhan 2021 [22]	BMI, fat mass, lean body mass, body fat percentage	BIA(InBody 770,Cerritos, CA)

BMI Body mass index, DXA Dual-energy X-ray absorptiometry, BIA Bioelectrical impedance analysis, N/R Not reported

BMI is a widely used and simple indicator of overall adiposity.Six arms from five RCTs reported changes in BMI(n = 184) [16–18, 21, 22]. The pooled analysis showed no significant improvement following HIIT (SMD=–0.16; 95% CI [–0.45,0.13]) with low heterogeneity. Subgroup analysis indicated no significant differences when HIIT was compared with MICT (SMD=–0.36; 95% CI [–0.98,0.27]) or with UC (SMD=–0.11; 95% CI [–0.43,0.21]). Sensitivity analyses confirmed the robustness of these findings(Fig. 8a).

Fig. 8.

Forest plot of pooled subgroup analysis of body composition. Forest plots showing pooled standardized mean differences (SMDs) with 95% confidence intervals for (a) body mass index, (b) fat mass, (c) lean body mass, and (d) body fat percentage

Fat mass reflects the absolute quantity of adipose tissue.Six arms from five studies reported this outcome(n = 190) [16, 17, 21, 22, 27]. Meta-analysis showed that HIIT did not significantly reduce fat mass (SMD=–0.23; 95% CI [–0.51,0.06]), with low‑to‑moderate heterogeneity (I²=34.8%). Subgroup analyses comparing HIIT with MICT (SMD=–0.41; 95% CI [–1.04,0.21]) and with UC (SMD=–0.15; 95% CI [–0.66,0.36]) likewise showed no significant differences. However, sensitivity analysis indicated that exclusion of Gonzalo‑Encabo 2022 resulted in a significant effect (SMD=–0.36; 95% CI [–0.67,–0.05]; P = 0.024), suggesting that individual studies may influence the overall result(Fig. 8b).

Lean body mass represents the total mass of metabolically active tissues, such as muscle, bone and viscera.Six arms from five studies assessed this parameter(n = 190) [16, 17, 21, 22, 27]. Overall, HIIT had no significant effect (SMD = 0.22; 95% CI [–0.06,0.51]) with low heterogeneity. Subgroup analyses showed no significant differences when compared with MICT (SMD=–0.05; 95% CI [–0.67,0.57]) or UC (SMD = 0.30; 95% CI [–0.02,0.63]). Sensitivity analyses confirmed the stability of these findings(Fig. 8c).

Body fat percentage is considered a more precise indicator of relative body composition.Five arms from four trials reported this outcome(n = 170) [17, 21, 22, 27]. Meta-analysis showed no significant reduction after HIIT (SMD=–0.32; 95% CI [–0.88,0.24]), with moderate‑to‑high heterogeneity (I²=69.2%). Subgroup analysis comparing HIIT with UC (SMD=–0.29; 95% CI [–0.99,0.41]) also indicated no significant difference, with high heterogeneity within the subgroup (I²=76.7%). Sensitivity analyses did not alter the overall conclusion(Fig. 8d).

IL‑6

IL‑6 is a key inflammatory cytokine involved in systemic inflammation and tumor progression.Four arms from two RCTs evaluated the effect of HIIT on IL‑6 levels(n = 63) [17, 26]. Meta-analysis showed no significant reduction in IL‑6 levels (SMD=–0.22; 95% CI [–0.65,0.22]), with low heterogeneity (I²=38.4%). Subgroup analyses comparing HIIT with UC (SMD=–0.20; 95% CI [–0.80,0.40]) and with MICT (SMD=–0.29; 95% CI [–1.61,1.04]) likewise showed no significant differences. The comparison with MICT showed higher heterogeneity (I²=77.2%), suggesting variability across studies(Fig. 9).

Fig. 9.

Forest plot of pooled subgroup analysis of IL-6. Forest plot showing standardized mean differences (SMDs) with 95% confidence intervals for the effect of high-intensity interval training (HIIT) on circulating interleukin-6 (IL-6) levels

Certainty of evidence

The certainty of evidence for the primary outcomes and the secondary outcomes was assessed using the GRADE framework.Detailed assessments are presented in Supplementary Tables S4 and S5.

Discussion

Advances in breast‑cancer therapy have markedly improved, but treatment‑related adverse effects remain a major challenge, impairing quality of life, body composition, cardiorespiratory fitness and immune function. Many patients experience fatigue, weight gain, metabolic disturbances, muscle atrophy and reduced cardiovascular performance during therapy [4]. These complications compromise not only physiological well-being but also social well-being, highlighting the need for efficient rehabilitation strategies. HIIT, which consists of brief bursts of intense exercise alternated with recovery periods, has gained attention for its time efficiency and potent physiological benefits.

Our meta‑analysis indicates that HIIT significantly enhances VO₂peak compared with UC and yields improvements comparable to MICT. VO₂peak, a core indicator of cardiorespiratory fitness, correlates positively with exercise tolerance, cardiovascular health, quality of life, treatment adherence and long-term survival, while low VO₂peak predicts higher cancer‑specific mortality [28].Mechanistically, HIIT improves both central and peripheral function. Centrally, it increases cardiac output and contractility via sympathetic activation. Peripherally, it activates the AMP‑PGC‑1α pathway, promoting mitochondrial biogenesis, oxidative enzymes, and capillary density, which enhance oxygen extraction and muscle metabolism [29]. Furthermore, HIIT mitigates anthracycline-induced cardiotoxicity by reducing oxidative stress, preserving mitochondrial dynamics, suppressing NF-κB–mediated inflammation, and inhibiting apoptosis- and fibrosis-driven remodeling. These integrated adaptations converge to preserve cardiac structure and function, reflected in improved VO₂peak, left ventricular ejection fraction, and other indices of cardiopulmonary performance [30, 31].

In the present analysis, no statistically significant difference in VO₂peak was observed between HIIT and MICT. However, previous meta-analyses in cancer populations have reported greater VO₂peak improvements with HIIT compared with MICT [32, 33], suggesting that discrepancies may stem from differences in patient characteristics, exercise protocols, limited physiological reserve, and small sample sizes. It is also possible that when MICT is prescribed at sufficient intensity or duration, it elicits near-maximal cardiovascular adaptations, narrowing the between-group differences. Nonetheless, these findings indicate that HIIT can achieve comparable VO₂peak gains to MICT while requiring less time commitment.

Improvements in VO₂peak and functional capacity can also translate into enhanced quality of life, a key indicator of holistic well-being during and after cancer treatment. Consistently, our findings show that HIIT may improve emotional, functional and social domains compared with usual care, whereas physical well-being showed no clear improvement. These findings suggest that HIIT may help patients regain social roles and confidence by rapidly improving physical capacity, thereby enhancing self‑efficacy and alleviating anxiety or depression [34]. The lack of benefit in the physical dimension may reflect the modest changes in body composition, as well as the need for longer duration to detect subtle effects.

Given that physical capacity and psychosocial well-being are closely linked to underlying body composition, it is also important to consider structural markers of health. Body‑composition indices such as BMI, fat mass, lean body mass and body fat percentage are strong prognostic markers in breast cancer.Reviews have shown that unfavorable body composition, characterized by low muscle mass and high fat mass, is associated with increased cancer-specific mortality and reduced overall survival across multiple cancer types [35–37].Our analyses showed that HIIT alone did not significantly improve these parameters compared with MICT.Although HIIT elicits strong acute metabolic responses, its limited duration likely restricts total energy expenditure. Combining HIIT with resistance training or dietary interventions may therefore be necessary to produce measurable changes [5]. Post-exercise compensation behaviors such as increased appetite could also offset fat loss, and notably, none of the included trials controlled for diet [38]. Excluding Gonzalo-Encabo 2022 made the overall fat-mass reduction significant, indicating its disproportionate influence. This trial was conducted during active anthracycline chemotherapy, unlike the others, and exercise adaptations in this phase may improve metabolic and cardiometabolic function without altering body composition. The dissociation between improved metabolism and unchanged fat mass suggests that its outlier status reflects phase‑specific physiological response rather than statistical or measurement bias.

Body composition assessment methods varied across studies. Skinfold measurements lack rigor, and BIA is less accurate than DXA, tending to underestimate fat mass and overestimate lean mass in breast cancer survivors [39]. Treatment-related factors can further confound measurements. Axillary lymph node dissection and radiotherapy may cause lymphedema, taxanes can increase capillary permeability, anthracyclines and HER2-targeted therapies may promote peripheral edema, and tamoxifen may induce mild fluid retention [4]. These fluid shifts affect both BIA and DXA, with BIA overestimating lean mass due to its sensitivity to whole-body water content, whereas DXA, although generally more robust, can still be biased in the presence of abnormal fluid accumulation such as pleural or ascitic effusions [40].

Inflammatory regulation represents a fundamental mechanism through which exercise influences tumor behavior and treatment outcomes.The absence of a significant reduction in resting IL-6 following HIIT in our meta-analysis should be interpreted within the clinical and biological context of the included populations. One trial involved postmenopausal women with overweight or obesity, predisposed to chronic low-grade inflammation and elevated adipose-derived cytokines [26, 41], while the other included non-metastatic breast cancer patients post chemo-radiotherapy receiving tamoxifen [17]. Prior reviews indicate that treatment-associated inflammatory alterations can persist for months, sustaining enhanced pro-inflammatory cytokine capacity [42].These differences in baseline inflammatory status suggest that the same exercise intervention may elicit distinct IL-6 responses depending on the underlying immune and metabolic environment.

IL-6 exhibits context-dependent, temporally dynamic effects. Acute exercise induces muscle-derived IL-6, which activates AMPK and downstream STAT3 in skeletal muscle and liver, enhancing glucose uptake, fatty acid oxidation, and mitochondrial biogenesis while promoting anti-inflammatory cytokines such as IL-10 and IL-1 receptor antagonist [43–45].In contrast, chronic IL-6 elevations in tumor or inflammatory microenvironments sustain STAT3 and NF-κB signaling, recruiting regulatory T cells and M2 macrophages to establish immunosuppressive niches [43, 46].These opposing roles illustrate IL-6’s integrated functions across metabolic, immune, and tumor-regulatory networks. Moreover, IL‑6 was measured only at a single time point after the intervention; thus, such transient benefits were likely not captured.

Adherence and safety are critical determinants of exercise efficacy. Across trials, HIIT achieved moderate-to-high adherence comparable to moderate-intensity training, indicating that higher intensity does not compromise participation. Adherence mediates dose exposure to physiological adaptations, particularly cardiorespiratory fitness, where sufficient volume drives central and peripheral changes underlying VO₂peak improvements. In contrast, quality-of-life benefits depend partially on engagement and psychosocial reinforcement, while body-composition changes are more limited by intervention duration, energy expenditure, and dietary factors than by adherence alone. Safety is a primary consideration for HIIT in breast cancer populations. Across ten trials (n = 330), HIIT showed a consistently favorable short-term safety profile, with no serious (grade ≥ 3) events and only transient, mild musculoskeletal discomfort reported in a few postoperative participants(Table S6). Although high intensity theoretically raises risks of musculoskeletal strain, cardiovascular stress, and endocrine or psychological perturbations, current evidence supports HIIT as feasible and well-tolerated when supervised or appropriately monitored.

In this analysis, HIIT yielded improvements comparable to MICT without demonstrating superiority. Its efficacy may stem from greater excess post-exercise oxygen consumption (EPOC), which sustains elevated energy expenditure and oxygen uptake to support phosphagen resynthesis, lactate clearance, thermoregulation, and hormonal normalization [47]. Consequently, HIIT can elicit metabolic demands similar to longer sessions of continuous exercise despite shorter duration and lower volume. These findings align with ACSM and international oncology exercise guidelines, which emphasize safety, feasibility, and clinically meaningful gains in cardiorespiratory fitness and quality of life, rather than superiority of a specific modality, highlighting HIIT as a time-efficient option for breast cancer rehabilitation.

Nevertheless, this meta-analysis has limitations. The included trials were small and of short duration, limiting statistical power and long-term interpretation. Publication bias was not assessed, as most outcomes included fewer than 10 studies, rendering formal tests unreliable. Blinding of participants and personnel was generally not feasible, which may have introduced performance bias. Differences in disease stage reporting prevented stage-specific analyses, and exercise adherence was inconsistently reported, potentially affecting the observed effects and may limit the generalizability of our findings. Larger, well-designed trials with longer follow-up and standardized outcome assessment are needed to confirm the long-term efficacy and safety of HIIT.

Conclusions

HIIT is a safe and time-efficient intervention that may improve cardiorespiratory fitness and may benefit emotional, functional, and social aspects of quality of life in patients with breast cancer, showing benefits over usual care. However, these quality-of-life findings are based on a limited number of studies and small sample sizes. No clear effects were observed on body composition, and evidence for the inflammatory marker IL‑6 was insufficient to draw a conclusion. HIIT did not show superiority over MICT. Integrating HIIT with resistance training and dietary management may further enhance physical recovery and psychosocial outcomes. Large, well-designed RCTs with standardized protocols and longer follow‑up are needed to confirm its long-term efficacy and guide clinical implementation.

Abbreviations

MeSH

Medical subject heading

HIIT

High-Intensity Interval Training

MICT

Moderate-Intensity Continuous Training

UC

Usual Care

PRISMA

Preferred reporting items for systematic reviews and meta-analyses

BMI

Body Mass Index

FACT-G

Functional Assessment of Cancer Therapy - General

SMD

Standardized Mean Difference

Authors’ contributions

Conceptualization: Jie Zheng and Rui Yan.Methodology: Jie Zheng, Rui Yan and Shuhang Zhou.Data curation: Jie Zheng, Shuhang Zhou, Jiale Lin.Formal analysis: Jie Zheng and Rui Yan.Investigation: Shuhang Zhou and Jiale Lin.Original draft writing: Jie Zheng, Rui Yan, Shuhang Zhou and Jiale Lin.Review and editing: Jie Zheng.Supervision: Jie Zheng.”

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.