Moderate Intensity Exercise Reduces Side Effects of Cancer Therapy and Maintains Cardiorespiratory Fitness in Male Breast Cancer Patients: Findings from the BRECA Male Crossover Study

Male breast cancer (MBC) is a rare and poorly studied disease [1‒4]. Recent epidemiological data from the World Health Organization (WHO) suggest that approximately 0.5–1.0% of breast cancer cases per year occur in men worldwide [5]. Of note, the number of newly diagnosed MBC cases per year has increased from approximately 8,500–23,100 over the past 20 years [6]. In contrast, female breast cancer (FBC) is the most diagnosed cancer worldwide (11.4%) according to GLOBOCAN cancer statistics 2020 [7]. Based on the extensive literature and clinical experience in women, treatment standards have been largely extrapolated to male patients [3, 8‒11]. Thus, while breast cancer is well-documented in females, it is important to note that occurrences of MBC are exceptionally rare. Consequently, due to the rarity of breast cancer in males, smaller sample sizes are anticipated. Also, data on different treatment protocols in men have not yet been carefully studied, which seem very important, given the evidence of sex-differences in epidemiology, risk factors, clinical presentation, survival rates, and pathology [3, 12‒18]. In approximately 40% of men, the disease is not detected until later stages 3 and 4, mainly due to the lack of appropriate screening tools [19]. Also, men are significantly older than women at the time of diagnosis (69.6 years vs. 61.7 years, p < 0.001), and in combination, these factors contribute to a poor prognosis, while survival rates appear to be worse in male compared to FBC patients [13, 14, 20, 21]. Importantly, MBC survivors, like other cancer patients, suffer from various side effects such as postoperative lymphedema, tamoxifen-related cardiotoxicity, and erectile dysfunction, leading to a reduction in quality of life (QoL) [22‒25]. Therefore, studies with supportive interventions need to be conducted to improve the overall QoL of BMC patients in a sustainable manner [26, 27]. In this context, supervised exercise interventions have been proposed as a viable countermeasure that mitigates many side effects of cancer therapies and improve cardiorespiratory fitness (CRF) in FBC patients [28], male prostate cancer [29], and colorectal cancer patients [30]. Some authors have even suggested that a higher CRF level and subsequently higher V̇O₂ peak is an independent predictor of overall survival in metastatic breast cancer in women [31]. Therefore, well-designed exercise interventions are a prerequisite for improving QoL and overall well-being in cancer patients and survivors. However, to date, no exercise studies have been conducted with MBC patients, so data on exercise prescription in this population are limited, and tailoring exercise interventions remains a challenge. Retrospective population-based case-control studies with small sample sizes [32] that assess physical health status or pooled data from case-control and cohort studies that describe leisure-time activity at baseline [33] have been conducted. However, there is a lack of randomized controlled trials (RCTs) on prescribing supervised exercise or specific physical activity interventions for this patient population. The evidence base for current guidelines and recommendations for physical activity in cancer patients comes primarily from FBC or prostate cancer trials [34]. For example, in their systematic review of physical activity in cancer prevention and survival, McTiernan et al. [35] advocate for future studies that include less common cancers. This highlights the importance of further studies in patients with rare entities, such as MBC. Therefore, we conducted the first study with exercise intervention in MBC patients. The aim was to investigate the feasibility and the potential effectiveness of moderate and vigorous web-based exercise intervention on CRF and self-reported side effects in MBC patients. We hypothesized that this exercise intervention would reduce cancer therapy-related side effects to a similar extent after moderate and vigorous exercise protocols, whereas greater improvements in CRF would be expected after vigorous exercise (as compared to moderate) in MBC patients.

The study protocol was approved by the Ethics Committee of the German Sport University Cologne, and the study was registered in the German Register for Clinical Studies (DRKS00006795). All study procedures and protocols were performed in accordance with Good Clinical Practice and the Declaration of Helsinki. Before the start of the study, all participants gave written informed consent to participate in the present study. The following inclusion criteria were established: male participants after primary breast cancer diagnosis, age ≥18 years, breast cancer treatment (chemotherapy and radiotherapy) competed 6 months ago, medical fitness statement/medical clearance to participate, internet access. Participants with metastatic disease and acute orthopedic or internal medicine conditions were excluded from the study. The RCT crossover design was chosen to promote internal validity of data because of the rarity of the disease and the resulting lower availability of patients. This design allows participants to act as their own control group. Using a computer-generated randomization scheme, all men (at baseline) were assigned to moderate and vigorous exercise intensity (see Fig. 1 for details). To be included in the data analysis participants were required to complete at least 80% of the recommended study training. No upper limit was set for additional training, as this was the first study to look at men after breast cancer therapy and the effects of exercise on their overall well-being.

The recruitment period was from 2014 to 2016, and participants were recruited via the “Netzwerk Männer mit Brustkrebs e.V.” Recruitment was also done digitally via the website of the German Sports University and the “Mamma Mia Brustkrebs-Magazin.” In addition, some breast cancer clinics were contacted by e-mail and some study flyers were displayed in oncology/radiological medical practices. A total of 26 men who had completed breast cancer therapy within the past 6 months volunteered to participate, of whom 22 men met the inclusion criteria and participated in the present study.

All data were collected at the University Hospital of Cologne at four different time points. More specifically, data were collected at the beginning of the study – baseline (t0), at the end of exercise intervention – after 3 months (t1), at the beginning of the crossover baseline – after 4 months (t3), and finally at the end of the intervention – after 7 months (t4), with schematic illustration given in Figure 1. To determine individual CRF, a graded exercise test was performed at each visit on a stationary ergometer (Ergoline 900, Hamburg, DE) connected to a metabolic cart (Cortex: Metalyzer^® 3B-R2), in accordance with the recommendations of WHO (modified WHO graded test, starting with 30 W and increasing the load by 15 W every 2 min until task failure). Before each laboratory visit, the metabolic cart was calibrated according to the manufacturer’s guidelines, whereas peak V̇O₂ data were obtained from the last minute of the cardiopulmonary exercise test (CPET) using MetaSoft Studio software. Peak power output (PPO) and maximum heart rate (HR) were defined as data obtained at the end of the test(s), while cycling cadence was set at 60–70 rpm. To assess cancer-specific side effects (fatigue, erectile dysfunction, etc.), participants were required to complete five questionnaires (Eortc-QLQ-C30/-BR23, MFI-20, IIEF, AMS, and GPAQ). More precisely, the previously validated and reliable EORTC-QLQ-C30 is the standard instrument for assessing QoL in cancer patients [36, 37]. The EORTC-module-BR 23 is used specifically for breast cancer patients, regardless of disease stage [38]. The MFI-20 is a multidimensional fatigue questionnaire and therefore differentiates between different fatigue dimensions (emotional, general, etc.). The validated IIEF questionnaire aims to assess “male sexual function” [39]. The Aging Male’s Symptom Scale is offered to men to assess their age-related QoL [40], while the GPAQ was developed by the WHO and is used to assess physical activity status [41].

In this study, only an online training program was conducted because in-person training was not considered feasible due to the low incidence of MBC in a specific region. Patients recruited here were distributed throughout Germany, and 1 patient was from Austria. The program used was the EmotionNet system, which has been shown to be effective in other similar studies [42]. It is an interactive online platform provided by medi train Gbr (Erlangen, Germany) to record physical activity. All participants were given access to two online diaries within the EmotionNet system. To ensure that participants followed the protocol, a sports therapist was given access to both diaries throughout the exercise intervention. This allowed the therapist to adjust each workout (e.g., modify the exercise, such as adding more repetitions to achieve the targeted RPE value, etc., if necessary). Additional comments, such as difficulties or general questions, could be submitted via “EmotionNet system,” by e-mail or by telephone. Thus, the training was supervised even if the sports therapist was not present at each training session, and participants were required to record their weekly study-related training (endurance and strength training). For endurance training, information such as average HR (Polar, Oy, Finland), type of training, distance, subjective exertion perception (RPE) readings, and comments (problems or difficulties) had to be collected and entered in the “movement diary.” In addition, all other physical activities outside the study were also recorded in this diary to obtain as much information as possible. Strength exercises were recorded in a second online diary, the “exercise diary,” on the same EmotionNet system platform. At the end of each exercise, all participants had to document their achieved RPE value. As described earlier, the target RPE readings at moderate intensity is 11 “fairly easy.” For intense training, the target is RPE 15, which corresponds to “hard.” The moderate intensity intervention consisted of endurance training at 40–50% HR max., and strength training was optimized via Borg/(RPE): 11: “fairly easy,” while the vigorous exercise intensity included endurance training at 70–80% HR max. and strength training with RPE: 15: “hard” during the first 3 months (t0–t1). After a 4-week washout period without any exercise intervention, participants switched training intervention models. Training was conducted three times per week. More precisely, twice a week, 30 min endurance training was performed based on the individual preferences of the participants (walking, inline skating, dancing, running, etc.). During the endurance training, the men had to wear a HR belt around their chest to follow their individual exercise intensity zone. The Karvonen formula was used to determine the individual HR goal for each participant [43]. Strength training (3 × 10 repetitions each and 30 s rest at a pace of 6 s) was performed once a week (squats, crunches, diagonal leg and arm raises [prone position], dynamic push-up on knees, dynamic toe stand exercise, deadlift with weights). These six selected whole-body exercises were maintained throughout the duration of the study.

Descriptive analysis of participants’ anthropometric characteristics and all parameters at baseline was performed (n = 14). The effects of the group and treatment on CPET parameters were analyzed using a general line model, with Greenhouse-Geisser correction which was appropriate. If a significant F test was determined, a Bonferroni post hoc was applied to determine multiple comparisons. The presence of a treatment effect was calculated using an unpaired samples t test. To rule out the possibility of a carry-over effect, the sums of the measured values from both time periods were used and calculated for each subject. Subsequently, both sequence groups were compared using a t test for independent samples [44]. The level of statistical significance was accepted at p level <0.05. SPSS version 26 (IBM, USA) was used for statistical analysis and Excel Microsoft version 2019 for graphs.

A total of twenty-six men volunteered to participate in our study. Of these, three participants did not meet fitness requirements and one had to be excluded due to metastases. Thus, twenty-two participants were eligible for study inclusion and randomization. Of these, fourteen men completed all study procedures (details given below) and were included in the final calculations. There were no significant differences between groups in anthropometric or medical treatments/history at baseline (t0) among all participants (Table 1). All men included in the final calculations reported having undergone surgery, and 92.8% of participants were receiving antihormone therapy (tamoxifen).

A total of eight participants left the study. More precisely, during moderate intensity, five men left the study; while during intense phase, three men left. After the first 3 months, the dropout rate was 23%. At the end of the study, after 7 months, 36% of participants dropped out, and the adherence rate was 64%. Reasons for dropout were lack of time (n = 1), relocation (n = 1), inpatient rehabilitation stay (n = 1), lack of motivation and interest (n = 1), loss to follow-up (n = 1), surgical procedures (orthopedic cosmetic, n = 2). One participant dropped out of the program because he felt the training assigned to him at moderate intensity was self-perceived as too low.

The CRF data are shown in Table 2. A total of 14 participants completed all study procedures. To be included in the final report, they had to achieve at least two maximal effort criteria during the CPET protocols, including a plateau in V̇O₂ uptake and >1.10 respiratory exchange ratio. The mean V̇O₂ peak at baseline for all participants was 28.14 ± 4.43 mL⋅kg·min⁻¹. In response to two different training interventions (moderate and intensive exercise models, respectively), no differences were found between groups in absolute V̇O₂, V̇CO₂, and PPO (Table 3). However, the HR max. was 10 beats lower (time effect, p = 0.003) at the end of the intense training, with no interaction effects (p = 0.110).

Significant differences after exercise intervention were found in Physical Function and Social Function items, both assessed with the EORTC-QLQ-C30 questionnaire (Table 3). More specifically, the intensive training led to an overall decrease in Physical Function (−6.67 ± 4.92), and no changes were observed after the moderate training (5.72 ± 7.17). The difference between the intensive and the moderate treatment effect was significant (p = 0.037). Social Function scores increased in general after moderate training (IG1: 9.76 ± 3.79). Also, intensive training resulted in a decrease in Social Function (IG1: −30.15 ± 22.55; IG2: −4.76 ± 2.82), suggesting that responses to treatments were indeed different (p = 0.016), at least for these self-perceived items. In other items of EORTC-QLQ-C30, no significant differences were found in response to exercise interventions at two different intensities.

In general, global QoL scores decreased by (IG1: −8.34 ± 7.22) and (IG2: −14.29 ± 14.02) after intensive training, whereas in parallel, moderate training resulted in an improvement in QoL in (11.51 ± 2.45) with no changes after exercise treatment (1.19 ± 3.37). There were no differences observed in self-reported fatigue (p = 0.306) or any other item of the EORTC-QLQ-C30 symptom scale in response to the training intervention at different intensities (Table 4). There was also no significant difference between treatment effects (p = 0.204). No differences were found in the items of the EORTC-QLQ-BR23 Function Scales (p > 0.05), although there was a trend toward improvement in the self-reported item “Breast symptoms” (p = 0.095) after moderate training. Finally, self-reported fatigue items derived from the MFI-20 general fatigue scale generally showed no change following the training interventions, likely due to the small sample size.

Our study fills a significant gap by introducing exercise intervention to MBC patients, an area with limited investigation. The web-based exercise interventions, at both moderate and vigorous levels, demonstrated practicality and potential effectiveness in addressing CRF and self-reported side effects. Notably, our findings indicate that moderate exercise interventions could alleviate specific treatment-related side effects. Importantly, the main findings can be summarized as follows: (i) no adverse effects were reported during exercise intervention, supporting the safety and feasibility of our exercise intervention; (ii) no significant differences in peak V̇O₂, HR max., or PPO reached during CPET were observed between treatments; (iii) significant treatment effects were found for self-reported items “Physical Function” (p = 0.037), and in favor of moderate intensity for “Social Function” (p = 0.016). To our knowledge, the BRECA male RCT is the first study to address the effects of exercise interventions by combining data from self-reported items and objective measures of CRF, to provide a more comprehensive insight into the overall QoL, well-being, and physiological adaptation to exercise in a unique population of MBC. This is particularly important since we were able to recruit (and train) participants nationwide using this web-based approach, thus solving the problems of long-distance travel and study adherence in MBC survivors. Overall, study adherence was satisfactory, with an overall dropout rate of 36% after 7 months of crossover intervention, while a total of 14 participants completed 80% of the prescribed moderate and vigorous exercise with no adverse effects reported. The data on dropout rate presented here (35%) are generally consistent with previously reported web-based, home-based, or rare cancer entity studies [42, 45‒47], although some authors [48] have pointed out that web-based exercise interventions are often characterized by high dropout rates. The dropout rate reported here is somewhat higher compared to the systematic review of web-based interventions by Kuijpers et al. [46], who reported 20% dropouts after ∼6 months of intervention. However, the overall methodological quality of the papers reviewed by Kuijpers et al. [46] was moderate, and the included studies were very heterogeneous, ultimately leading to mixed results. To evade this methodological obstacle, we organized a nationwide RCT with a crossover design, which resulted in a slightly higher dropout rate. This was a compromise we made to ensure the internal and external validity of the data presented in this study.

Here, no changes in peak V̇O₂ or other CPET-derived variables were observed in response to two different exercise intensities (Table 2). Results on CRF adaptations during the course of this study are to some extent inconsistent with the hypothesis of the study, suggesting that intensive training would result in a greater increase in V̇O₂ uptake compared with moderate training in MBC. This is particularly noteworthy because peak V̇O₂ uptake is considered an important, clinically relevant physiological indicator of functional capacity in the general population [49], and an independent predictor of overall survival after a cancer diagnosis [50]. Here, we found no changes in peak V̇O₂ (mean values of ∼28 ± 7 mL ⋅kg·min⁻¹) over the course of the study, and this is inconsistent with most previous work on FBC or other cancer survivors [51‒56]. However, our data on CPET in male patients are presented for the first time in the literature, which seems very important given the wealth of literature focusing exclusively on FBC and exercise, and the evidence of sex-differences in cardiovascular response to exercise [57]. The baseline peak V̇O₂ data reported here are ∼10% higher compared with the average normative reference values of healthy men in North America (∼23 mL⋅kg·min⁻¹) in the age group between 60 and 69 years [58]. This opens up the possibility that further increases in V̇O₂ peak were rather limited, in part due to the age of the participants in our study (MBC patients here were 60 years old). Indeed, cardiovascular plasticity deteriorates with age, limiting O₂ transport, and thus peak V̇O₂ during exercise [57]. Thus, it is reasonable to hypothesize that maintenance, rather than improvement of peak V̇O₂ was observed, likely due the age of the participants included here. However, this cannot entirely explain the lack of changes in CPET performance. We point to the fact that 92% of the MBC patients included in this study received hormone therapy (tamoxifen treatment for several years), which is linked to changes in body composition, lower energy expenditure, osteopenia/osteoporosis, and may even be associated with cardiovascular events [59]. Therefore, our results on V̇O₂ peak are to some extent in line with recent case study of 28-year-old elite female athletes diagnosed with stage 1A breast cancer [60]. Indeed, data presented by Pérez-Bilbao et al. [60] showed a 12% decrease in peak V̇O₂ (from 55 before diagnosis to 48 mL⋅kg·min⁻¹) during a 14-month follow-up period, while taking tamoxifen, and despite her intense training during the follow-up period. During this 14-month period, her red blood cell and hemoglobin concentrations decreased by approximately 10%, which would explain, at least in part, why this young elite athlete taking tamoxifen was unable to bring her CRF level or power outputs to pre-diagnosis levels despite intense exercise. These insights offer a foundation for tailoring interventions and refining rehabilitation strategies specifically for MBC patients, thereby enhancing their overall well-being. Further studies should examine the sex-differences in BC patients and the possible effects of prolonged hormone treatments on energy metabolism during exercise. The study by Mijwel et al. [61] demonstrated that HIIT during chemotherapy improves several self-reported treatment side effects, such as fatigue, QoL, and emotional functioning for FBC patients. Additionally, it even contributes to reducing societal costs linked with extended sick leave.

Although no changes were observed in CRF measures in response to training intervention, the self-reported item of Social Function improved (by 67% on average) after moderate training and decreased by 35% after intensive training, as indicated by the significant difference in treatment effects (p = 0.016). These clinically relevant differences could be due to the fact that the moderate training was performed in a group setting. To support this explanation, work of Eime et al. [62] and Emslie et al. [63] showed that group activities have a better effect on social function, when compared to exercise performed autonomously. Further, Ammitzbøll et al. [64] conducted a RCT on strength training after breast cancer treatment in women and found that the psychosocial component of group exercise may be more beneficial than intensity for improving Social Function. However, it remains controversial whether this improvement in Social Function item is solely explained by the group training. In parallel, a similar trend was observed here in Physical Performance, where an 8% increase was observed after moderate training, followed by a 5% decrease after intensive training in both groups. Such results are not entirely in line with the current literature on cancer, exercise, and self-perceived items of exercise intensity. More precisely, Campbell et al. [34] suggested in the ACSM Exercise Guidelines for Cancer Survivors that self-reported Physical Function improves significantly after 8–12 weeks of aerobic and/or resistance training (or a combination of both), regardless of the exercise intensity, while supervised training appears to be more efficient compared with unsupervised home-based interventions. Improvements in self-reported Physical Performance following low-volume HIIT training were also reported by Reljic et al. [65]. These improvements were observed exclusively with low-volume HIIT training (5 × 1 min, >80% of HR max., with 1 min rest in between), whereas no improvements were observed in low-intensity controls SHAM. Thus, the role of exercise intensity and its effects on self-perceived items requites further investigation. Another valid explanation for the general lack of differences in self-reported items, including Role function, Emotional function, and Cognitive function items was the relatively high baseline scores at baseline. Baseline scores in all five items for the BRECA population were somewhat higher or very similar compared with the general population in Germany and worldwide [66, 67]. Therefore, there was apparently little room for improvement, and the need to improve these scores was not urgent, as this group of patients was unlikely to feel physically limited in their daily functions. Overall, the data of LoQ presented here are generally in line with the current literature on exercise, cancer and perceived LoQ [34, 68‒70]. Interestingly, a non-significant trend toward improvement in self-reported “Breast symptoms” (p = 0.095) was observed after moderate training, likely due to a small sample size. For the remaining self-reported items, the three sub-scales (nausea/vomiting, sexual functioning, and side effects of systemic therapy) showed no change in either group after the moderate and intensive training. Also, no changes were observed in self-reported fatigue probably due to time elapsed since diagnosis and completion of chemotherapy and radiotherapy, these parameters appear to be unaffected.

While the relatively small sample size could be perceived as a potential limitation, it is important to consider the possibility that MBC may be an infrequent condition, potentially contributing to the absence of discernible differences between treatments. Hence, it is advisable to interpret these findings cautiously, especially given the 35% dropout rate, which amplifies the risk of inadequate statistical power (e.g., elevated β-error probability). The second limitation of our work is the lack of information on the inclusion of wearable technologies that could provide more comprehensive insight into physical activity levels during exercise protocols.

This was the first RCT with crossover design in MBC patients that confirmed the feasibility and effectiveness of exercise intervention at different intensities. This approach allowed us to address the problem of long-distance travel and study adherence in a unique population dealing with a very rare type of cancer. The main findings suggest that moderate intensity exercise training attenuates many of the side effects of cancer treatment, especially for more symptomatic less powerful participants whereas no differences between peak V̇O₂ levels suggest that both exercise protocols are suitable for maintaining CRF, even in aging cancer patients. Further studies should examine sex-specific differences in BC patients and the potential effects of hormone treatments on changes in energy metabolism. This could then be a useful tool for developing targeted, sex-specific exercise programs to improve and maintain overall health and well-being after BC in clinical (rehabilitation) and follow-up (cancer sports groups) settings.

This study was approved by the Ethics Committee of the German Sport University Cologne, and the study was registered in the German Register for Clinical Studies (registration number: DRKS00006795). All study procedures and protocols were performed in accordance with Good Clinical Practice and the Declaration of Helsinki. Informed consent was obtained from all individual participants included in the study.

The authors have no conflicts of interest to declare.

Funding was provided by Susan G. Komen Deutschland e.V.- Verein für die Heilung von Brustkrebs and by an anonymous private donation.

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by E.B.S., J.R., H.B., F.T.B., D.Z., and W.B. The first draft of the manuscript was written by E.B.S., D.Z., and F.T.B. commented on previous versions of the manuscript and read and approved the final manuscript.

The datasets generated during and/or analyzed during the current study are not publicly available due to the patient’s privacy/ethical restrictions but are available on request from first or leading author or the manuscript.