Five-year follow-up of the OptiTrain trial on concurrent resistance and high-intensity interval training during chemotherapy for patients with breast cancer

The protocol predefined aim of this study is to assess sustained effects of the OptiTrain trial on several health outcomes, 5 years after the baseline assessment. The OptiTrain study was a prospective, randomised controlled trial with 240 patients with breast cancer undergoing adjuvant chemotherapy that compared the effects of 16 weeks of two exercise programs, RT-HIIT and AT-HIIT, with usual care (UC). After a 5-year follow-up, eligible participants were evaluated for the primary outcome of cancer-related fatigue (CRF) and secondary outcomes including quality of life, symptoms, muscle strength, cardiorespiratory fitness, body mass, physical activity, and sedentary behavior. Statistical analysis was conducted using linear mixed models adjusted for baseline values. Tumour profile and menopausal status were additionally adjusted for CRF. Mean differences (MD), 95% confidence intervals (CIs), and standardized effect sizes (ES) were reported. At the 5-year follow-up, there were no statistically significant differences in total CRF between the intervention groups and the UC group. RT-HIIT reported significantly reduced pain sensitivity at the gluteus MD = 79.00 (95% CI 10.17, 147.83, ES = 0.55) compared to UC. Clinically meaningful differences for an increase in cognitive CRF and cardiorespiratory fitness were observed for the AT-HIIT versus UC group, and for lower limb strength for the RT-HIIT versus UC group, albeit without statistical significance. Engaging in targeted exercise during adjuvant chemotherapy for breast cancer provides short-term benefits in reducing fatigue and maintaining physical function. However, our 5-year follow-up indicates that these effects are limited in the long term. This underscores the need to support breast cancer survivors maintain their PA levels throughout their survivorship journey.


Outcome measures
An overview of the outcome measures used at the various assessment time points of the OptiTrain RCT is presented in Table 1.The primary outcome was cancer-related fatigue, and the secondary outcomes were symptoms and symptom burden, health-related quality of life, cardiorespiratory fitness, lower and upper body muscle strength, pressure pain threshold (PPT), objective measures of sedentary behaviour and physical activity, body mass, sick leave, heart failure diagnosis and cognitive functioning.Most outcomes were assessed at all time points: baseline (1 week before the second chemotherapy session), post-intervention (16 weeks after the baseline), 1-year post-baseline, 2 years post-baseline, and 5 years post-baseline.The exceptions to this were objectively measured PA which was assessed at baseline, 2 years, and 5 years only; the assessment of pressure pain threshold (PPT) that occurred at baseline, 16 weeks, 2 years, and 5 years only; and heart failure diagnoses and objectively measured cognitive functioning which were measured only at 5 years.
More comprehensive information regarding the methods for the outcome measures has been presented in previous publications [14][15][16][17] .Participants were asked to complete a single-item questionnaire on their sick leave status, with five options: 0% (no sick leave), 25%, 50%, 75%, or 100% (full-time sick leave).Information on heart failure (HF) diagnosis was obtained from the Swedish Heart Failure Registry 18 .Cognitive functioning was measured using the Amsterdam Cognition Scan (ACS), an online validated neuropsychological test battery 19,20 , which participants completed on their own computer at home.An overview of ACS tests, outcome measures and corresponding cognitive domains are found in Table 2.

Supervised exercise program
The 16-week OptiTrain exercise intervention has been described in previous publications [14][15][16][17] .In summary, both groups trained twice a week on non-consecutive weekdays for 16 weeks.Each session lasted approximately 60 min and was carried out at the exercise clinic at Karolinska University Hospital.An exercise physiologist or oncology nurse supervised all sessions to ensure safety, correct technique, and adherence to the exercise protocols.Each exercise session started with a 5-min warm-up on a cycle ergometer or treadmill at a Borg scale rate of perceived exertion (RPE) of 10-12 21 and concluded with a 10-min cool-down involving dynamic muscle stretching.
The RT-HIIT group performed eight resistance exercises targeting major muscle groups using using weight stack training equipment, participants' body mass, free weight dumbbells or barbells.Exercises included leg presses, bicep curls, squat jumps, tricep extensions, lunges, bench presses, sit-ups or Russian-weighted abdominal twists, shoulder presses, and prone-lying back extensions.Participants started with 2 sets of 8-12 repetitions at 70-80% of their estimated one-repetition maximum (1-RM).The RT-HIIT sessions concluded with three 3-min bouts of HIIT at a perceived exertion rating (RPE) of 16-18, with one minute of recovery between bouts on a cycle ergometer.
The AT-HIIT group began each session with 20 min of moderate-intensity (RPE 13-15) continuous aerobic exercise, followed by the same HIIT regimen as the RT-HIIT group.

Follow-up period
After the 16-week exercise program, participants received written prescriptions for both aerobic and resistance exercises.One-on-one counseling sessions with a health educator were offered, along with reduced-rate gym memberships.Over the 2-year follow-up (2014-2017), participants attended seven educational sessions covering topics like healthy lifestyles, fitness, and exercise programs to promote motivation and continued engagement.

Usual care
The control group (UC) received care as usual for breast cancer and written information on physical activity at the start of the intervention period, including exercise recommendations for cancer patients based on American College of Sports Medicine guidelines 11 , but no supervised exercise training or specific exercise prescriptions.

Sample size calculation
The initial power calculation used total fatigue (Piper Fatigue Scale) as the primary outcome after the 16-week intervention.With an effect size of 0.53 and a desired power of 0.8, 65 participants per group were needed.To account for a 20% anticipated dropout rate, 80 participants were recruited for each group.

Statistical analysis
Baseline demographics were summarized for all participants, including those who remained and those who dropped out.Summary statistics including means, standard deviations, frequencies, and percentages were utilized to conduct a descriptive analysis.Between group differences were calculated using linear mixed model (LMM) analysis using outcome measurements collected at 16 weeks (post-intervention), 1 year, 2 years, and 5 years after the baseline assessment adjusting for baseline values for all variables.For the primary outcome of CRF, tumour profile and menopausal status were also additionally adjusted for in the analysis based on knowledge of known prognostic factors.At the 5-year follow-up, completers were defined as those participants who had at least one non-missing response to the questionnaires at the 5-year assessment.For cardiorespiratory fitness outcomes, measured as predicted peak oxygen uptake (VO2 peak ) estimated VO2 peak (L⋅min −1 ) and estimated VO2 peak (mL⋅kg −1 ⋅min −1 )], measurements above physiologically acceptable cut-off limits of 65 mL⋅kg -1 ⋅min −1 ) were removed from the linear mixed models (N = 6; 5-year follow-up = 3; 2-year follow-up = 2; 1 year-follow-up = 1) considering also if the changes from the previous timepoints are significantly large to be acceptable.The outcome objectively assessed PA and sedentary behavior were analysed using validated wear-time specifications and cutoffs for adults 22 .Between-group differences in cognitive functioning 5 years after baseline were assessed using one-way ANOVA, after excluding outliers for ACS outcomes.For time-based outcomes (Connect the dots I and II, Reaction speed, Fill the grid), the median absolute deviation (MAD) 23 method was used for outlier detection.
The MAD was applied times 3.5 per age group (< 50 years, 50-59 years, ≥ 60 years).For count outcomes (Wordlist learning, Box tapping, Digit sequences I and II), 0 scores were excluded as these most likely reflect technical or human errors.Scores on Wordlist recall and recognition were excluded if the retention interval was longer than 45 min.Mean differences and 95% confidence intervals (CIs) were reported, along with standardized effect sizes (ES) that were calculated and interpreted as described previously 24,25 .A significance level of p < 0.05 was used for all statistical analyses.All statistical analysis was performed using the statistical software IBM SPSS Statistics Version: 28.0.1.1 (14).

Results
The flow diagram (Fig. 1) illustrates the progress of all participants throughout the study.At the 5-year postbaseline timepoint, a total of 179 survivors were eligible and invited to take part in the assessments.Among those, 95 participants (N RT-HIIT = 35; N AT-HIIT = 36; N UC = 24); 40% of the initially randomized individuals and 46% of those who completed baseline testing) answered the questionnaires, while 97 participants (N RT-HIIT = 33; N AT-HIIT = 34; N UC = 30; 40% of the initially randomized individuals and 47% of those who completed baseline testing) underwent at least one of the in-clinic physiological assessments.Nearly, 67 participants completed cognitive testing (N RT-HIIT = 24; N AT-HIIT = 27; N UC = 16).Among those who were eligible and invited, despite not participating in the assessments at the 2-year follow-up, 10 additional participants returned for the 5-year assessments.About, 68 participants who answered the questionnaires at the 2-year follow-up did not answer the 5-year questionnaires and were therefore considered dropouts.
Socio-demographic and clinical characteristics of the participants at baseline of the OptiTrain RCT and those who completed the 5-year follow-up and included in the present study are provided in Table 3.No statistical test was performed to compare these characteristics as they may necessarily not indicate the presence or absence of confounding factors or biases 26 .Similarly, characteristics of those who answered the 2-year follow-up but not the 5-year questionnaires is presented in Table 4 (no statistical test were performed to compare these characteristics).At the 5-year follow-up, a higher proportion in the RT-HIIT group had higher MVPA (90.8 ± 34.9), higher education (77.1%) and less frequent triple negative breast cancer (5.7%); the AT-HIIT group were more often post-menopausal (77.8%); a higher proportion in the UC group were married/partnered (79.2%) (Table 3).Besides this, a higher percentage among all three groups at the 5-year follow-up received taxane-based therapy compared to all participants and the dropouts (Tables 3, 4).In the Swedish Heart Failure Registry, HF diagnoses were registered for four study participants.Moreover, two additional cases of HF were diagnosed during the follow-up assessments of the RCT.
The attendance rates for the supervised exercise intervention were 68% for the RT-HIIT group and 63% for the AT-HIIT group.Regarding adherence to the exercise prescription in the supervised exercise intervention, it was 83% for the RT-HIIT group and 75% for the AT-HIIT group.On average, the attendance for the seven motivational seminars was twenty percent, with a range of 11% to 27%.The between-group differences of the mean scores (mean differences = MD) for the main outcome of CRF as assessed by using PFS is presented in Table 5.At 5 years, no group showed an average total fatigue severity cutscore above 3 (mild), and no statistically significant nor clinically meaningful differences were observed for total cancer-related fatigue between either of the intervention groups and the UC group; RT-HIIT vs UC: MD = − 0.02 (95% CI − 1.50, 1.46, ES = 0.10) and AT-HIIT versus UC: MD = 0.14 (95% CI − 1.34, 1.62, ES = 0.43).Similarly, no statistically significant differences were found for the four sub-scales of the PFS between either of the exercise groups and UC (Table 5).However, clinically meaningful differences for an increase in cognitive CRF MD = 0.38 (95% CI − 1.13, 1.88, ES = 0.68) were noted for AT-HIIT compared to UC.

Cardiorespiratory fitness, muscle strength, pain sensitivity, and body mass
No statistically significant differences were found between either exercise group and UC for cardiorespiratory fitness.However, clinically significant differences favouring the AT-HIIT group were found for both estimated VO2 peak (L⋅min -1 ): 0.18 (95% CI − 0.12, 0.47, ES = 0.55) and estimated VO2 peak (ml⋅kg -1 ⋅min -1 ): 1.62 (95% CI − 3.57, 6.82, ES = 0.54) (Table 8).Similarly, although there were no statistically significant differences between RT-HIIT and UC for lower limb muscle strength, clinically meaningful differences were seen favouring RT-HIIT: MD = 10.28,(95% CI − 2.83, 23.40, ES = 0.52).No statistically significant differences were observed for hand grip strength or body mass (Table 8).While no statistically significant differences were found for pain sensitivity assessed with PPT at the trapezius between the groups, there were statistically significant and clinically meaningful differences between RT-HIIT and UC favouring RT-HIIT for PPT at the gluteus: MD = 79.00(95% CI 10.17, 147.83,ES = 0.55).No such differences were found for AT-HIIT vs. UC in gluteal PPT.

Sedentary behaviour and physical activity
There were no statistically significant differences observed between either of the exercise groups and the usual care group in terms of minutes of sedentary behavior or moderate to vigorous PA (Table 8).Although not  www.nature.com/scientificreports/statistically significant, all three groups decreased their MVPA compared to the 2-year assessment point returning to their levels at baseline (Table 8).

Sick leave
At 5 years, as many as 97%, 86%, and 100% reported that they were not on any sick leave among RT-HIIT, AT-HIIT and UC respectively.Only 1 participant (3%) in the RT-HIIT group reported being on full-time sick leave and 3 participants (8%) in the AT-HIIT group reported being on some level of sick leave, among which 1 participant was on 25% sick leave and 2 were on 50% sick-leave.

Cognitive functioning
No statistically significant differences were found between the two intervention groups and the UC group for any of the cognitive tests (Table 9).

Discussion
This study examined the lasting impacts of the OptiTrain exercise program on breast cancer survivors 5 years post-intervention.The RT-HIIT group showed significant improvement in gluteal PPT compared to the control group.Clinically meaningful improvements were also seen for lower limb strength for the RT-HIIT and cardiorespiratory fitness for the AT-HIIT compared to UC group, albeit without statistical significance.Clinically significant increase in cognitive CRF was seen among the AT-HIIT group vs the UC.Evidence shows that breast cancer survivors experience several long-term functional changes, such as reduced strength, aerobic capacity, and mobility, compared to their peers in the same age group 27 .Moreover, despite generally comparable QoL to the general population 5 years post-diagnosis, many disease-free breast cancer survivors still faced adverse effects, necessitating ongoing screening and support for fatigue, sleep, cognitive issues, pain, menopausal/sexual symptoms, physical performance, and weight problems during and after therapy 28 .
However, in the present study a clinically relevant difference for cognitive CRF sub-scale favoring the UC group was observed in the AT-HIIT group compared to UC at 5 years but lacked statistical significance.These results are in contrast with the previous findings 14,17 of the OptiTrain trial where cognitive CRF was significantly lower among the intervention groups (AT-HIIT at the 1-year and RT-HIIT at the 2 year-follow-up mark).In fact, the fatigue scores for the total CRF and the four fatigue dimensions in the control group were lower than their baseline scores, that possibly imply a generally low level of fatigue.Moreover, none of the 3 allocation groups had an average fatigue severity score above 3, implying little remaining fatigue.Given that fatigue is a persistent side effect of cancer treatment, these are promising results for sustainable long-term effects.
No differences in cognitive functioning and subjective cognitive functioning were found between the exercise groups and the UC group at 5 years.These findings are in line with other studies that also found no long-term beneficial effects of PA on objective cognitive functioning 29,30 .Because cognitive functioning was assessed only at 5 years, we cannot correct for possible baseline differences between groups nor draw any conclusion about change in cognitive functioning over time.An ongoing follow-up study assessing cognitive functioning at baseline prior to neoadjuvant chemotherapy and 1 year later, will allow further investigation on whether physical exercise during chemotherapy for breast cancer could have any potential protective effect on cognition 31 .
At 5 years, the AT-HIIT group demonstrated clinically relevant differences in cardiorespiratory fitness compared to the control group but lacked statistical significance and may be attributed to lack of statistical power with only 40% of those who were randomised from the OptiTrain participants remained at 5 years.These  [15][16][17] and other studies that show that aerobic training results in sustained cardiorespiratory fitness 32 .Similarly, the RT-HIIT group still showed clinically www.nature.com/scientificreports/meaningful differences for leg muscle strength again without statistical significance potentially owing to the slightly diminished statistical power.Muscle strength is vital for patients, as it is linked to cancer-related fatigue and overall survival in older breast cancer survivors 33,34 .Furthermore, it's more critical than muscle mass in conditions like sarcopenia, which heightens the risk of falls, reduces independence, lowers quality of life, and  35 .Though additional research is required to confirm our initial results, suggesting that engaging in specific exercise during chemotherapy could be a promising way to improve cardiorespiratory fitness and lower body muscle strength for up to five years, these findings warrant attention.We speculate that these may partly result from the lasting habit formation and knowledge gained during the supervised exercise intervention.Participants in the RT-HIIT group reported favourable statistically and clinically significant differences in PPT gluteus than the UC group.These results are particularly promising as a higher percentage of participants at the 5-year assessment had received taxane-based therapy as their treatment regimen compared to the previous assessment points of the OptiTrain study (Table 3).Taxane-based therapy is now crucial in treating early-stage and advanced cancers, including breast carcinoma, and has led to increased survival rates 36 .Over the past decade, it has become widely used in breast cancer management.However, it comes with side effects like peripheral neuropathy, myelosuppression, joint and muscle pain, and skin reactions, which can negatively affect participants' quality of life [37][38][39][40] .Particularly, peripheral neuropathy can be severe in patients undergoing taxane therapy which may increase numbness, leading to decreased PPT 38,40 .This is promising because our prior findings at 16 weeks showed that the RT-HIIT group on taxane treatment did not become more sensitive to pain over time 15 .It's particularly encouraging that even 5 years after completing the exercise program, they continue to experience improved pain sensitivity in the gluteus, a significant muscle group for the lower extremities.Our study is the first to show that participating in specific exercise during chemotherapy for breast cancer shows promise as an effective method to reduce pain sensitivity in the gluteus even up to 5 years after the intervention.Despite the lack of statistical link between changes in gluteus PPT and lower-limb muscle strength (data not shown), evidence suggests a correlation between muscle strength and PPT in both healthy and diseased populations 41,42 .The RT-HIIT group's continued resistance to increased pain sensitivity at the gluteus, but not the trapezius, may www.nature.com/scientificreports/be due to their sustained higher lower-limb muscle strength, which remained significantly greater even five years post-intervention.We speculate that this difference may be partly due to the long-term benefits of exercise habit formation and knowledge gained from the supervised resistance exercise program.At the 5-year point, no significant differences in PAL were observed among the three groups.Moreover, all three groups reduced their MVPA levels compared to the assessment at 2 years, returning to baseline levels.In this study only 2 participants (3%) met the current PA recommendations of 150 minutes of MVPA per week.This suggests a potential decline in the long-term exercise adherence effects and habits developed during the supervised resistance exercise program, likely due to the absence of extended motivational support beyond the 2-year assessments to foster exercise autonomy among participants and the attendance rate at the motivational seminars up to the 2-year follow-up was already low.At the 5-year mark, most participants in all three groups self-reported a moderate activity level (57% in RT-HIIT, 58% in AT-HIIT, and 58% in UC, data not shown).Our findings contrast with those of the PACT study and the study by Mutrie et al, where the exercise intervention groups showed higher PAL four-and five-years post-intervention, respectively 6,7 .It is noteworthy that the methods for measuring PA differed between the studies (questionnaires vs accelerometers), making direct comparisons challenging.
Notable strengths of the present study include the long follow-up period with prospective data collection and the use of validated questionnaires.While prospective data collection minimizes selection bias, some potential bias might remain due to a reduced sample size of 95 participants at the 5-year assessment, mostly due to participant attrition.In experimental studies, where costs often constrain sample sizes, having 95 individuals for long-term follow-up can still be considered an acceptable attrition rate.Another strength is the use of accelerometers for objective, long-term PA and sedentary behavior measurement in a randomized controlled trial lasting up to 5 years that overcomes issues associated with self-reported questionnaires 43 .However, it's essential to recognize that accelerometers may not fully capture certain activities like resistance exercise, cycling, or water sports, particularly when worn on the hip.Additionally, a limitation of the study is the lack of specific details about the types of physical activities participants engaged in during the follow-up, potentially leading to an underestimate of overall PA levels, particularly during assessments.

Conclusions
Notably in our study, the RT-HIIT group exhibited substantial enhancements in gluteal PPT compared to UC.Additionally, clinically significant improvements were observed in lower limb strength for RT-HIIT and cardiorespiratory fitness for the AT-HIIT compared to UC, but an increase in cognitive CRF although statistical significance was not attained.However, it also highlights a concerning decline in physical activity levels over time.These findings provide new perspectives on the long-term benefits of incorporating an exercise program during chemotherapy for women in the early-curative setting with breast cancer.Nonetheless, the study also emphasizes the importance of developing enhanced strategies to assist survivors in maintaining their PAL and offering continuous motivational support and PA prescriptions support throughout their survivorship journey.

Figure 1 .
Figure 1.CONSORT flowchart illustrating the participant progression throughout the OptiTrain study.

Table 1 .
Outcome measures used in OptiTrain RCT at baseline and at 16 weeks, 1 year, 2 year and 5 years follow-up post-baseline.*Primary outcome; **Secondary outcomes.

Table 2 .
Cognitive tests of the Amsterdam cognition scan.

Table 3 .
Baseline characteristics of all OptiTrain participants and completers 5-years post-baseline.Continuous variables are presented as mean ± SD, whereas dichotomous or categorical variables are presented as %, RT-HIIT resistance training and high-intensity interval exercise group, AT-HIIT moderate-intensity and high-intensity interval training group, UC usual care, BMI body mass index, MVPA objectively measured moderate-to vigorous-intensity physical activity, SED objectively measured sedentary behaviour, HER + /human epidermal growth factor receptor 2 positive ( +) or negative (-), ER + /-estrogen receptor positive ( +) or negative (-), SD standard deviation.

Table 6 .
Symptoms and symptom burden 5 years post-baseline.SD standard deviation, CI confidence interval, ES effect size, RT-HIIT resistance training and high-intensity interval exercise group, AT-HIIT moderate-intensity and high-intensity interval training group, UC usual care, ES effect size.*p < 0.05.

Table 7 .
Health-related quality of life 5 years post-baseline.SD Standard deviation, CI Confidence interval, ES Effect size, EORTC European organisation for research and cancer treatment Quality of life questionnaire, RT-HIIT Resistance training and high-intensity interval exercise group, AT-HIIT Moderate-intensity and highintensity interval training group, UC Usual care, ES Effect size.*p < 0.05.

Table 8 .
Cardiorespiratory fitness, muscle strength, pain pressure threshold, physical activity, sedentary behaviour, and body mass 5 years post baseline.SD Standard deviation, CI Confidence interval, ES Effect size, RT-HIIT Resistance training and high-intensity interval exercise group, AT-HIIT Moderate-intensity and highintensity interval training group, UC Usual care, MVPA Objectively measured moderate-to vigorous-intensity physical activity, SED Objectively measured sedentary behaviour.*p < 0.05.

Table 9 .
Cognitive functioning 5 years post-baseline.SD Standard deviation, CI Confidence interval, RT-HIIT Resistance training and high-intensity interval exercise group, AT-HIIT Moderate-intensity and high-intensity interval training group, UC Usual care.a Higher scores indicate worse performance.