Five adverse events were recorded among four participants, all classified as grade 1: three cases of fatigue and two cases of muscular pain in the extremities. None prevented continued participation, and no serious adverse event occurred. The trial was too small and recruitment fell below target, so its clinical results remain exploratory and do not show whether exercise improved fitness, quality of life, fatigue, complications or hospital stay.

The study, known as the PReOperaTive Exercise Therapy for patients with cancer, or PROTECT, study, recruited adults scheduled for surgery at least two weeks away and medically cleared for exercise. Patients with metastatic disease, absolute contraindications to exercise testing or an adverse event during initial testing that made participation unsafe were excluded. The two-arm trial was registered prospectively and used a computer-generated allocation sequence in a 1:1 ratio, with participants stratified by primary cancer diagnosis.

Although an effectiveness objective and a non-inferiority sample-size calculation formed part of the original registration, the researchers shifted the report’s emphasis to feasibility and safety after the planned recruitment target was not reached. That revision was based on the recruitment shortfall, not on the direction or magnitude of the clinical results. Participants and intervention staff could not be blinded, while outcome assessors and data analysts were also unblinded.

The reported findings therefore concern whether a limited programme could be delivered safely and consistently, rather than whether it is effective. Prehabilitation uses the interval between diagnosis and surgery to prepare patients for the physiological stress of an operation. The PROTECT intervention consisted of one-on-one, on-site aerobic training delivered by an Accredited Exercise Physiologist two or three times per week until 48 to 72 hours before surgery.

Each session lasted 45 to 60 minutes, including a warm-up and cool-down, with at least 30 minutes of continuous cycling on an electronically braked cycle ergometer. Training began at 40% of heart-rate reserve, calculated from resting and maximum heart rate measured during a baseline cardiopulmonary exercise test. When patients adapted and no longer reached the lower intensity anchors confirmed by heart rate and perceived exertion, the target could be increased by 10 percentage points of heart-rate reserve every three sessions.

Intensity remained between 40% and 70% of heart-rate reserve, corresponding to ratings of perceived exertion from 11 to 16. Individual delivery was initially used because few patients were enrolled concurrently and was retained because of COVID-19 restrictions. Recruitment proved to be a major limitation of the single-centre study.

During the recruitment period, 312 patients with breast, colorectal or prostate cancer underwent surgery with curative intent at the study centre, but only 47 were identified as potentially eligible and approached for formal screening. Of those, 23 enrolled and were randomised: 14 to prehabilitation and nine to usual care, a recruitment rate of 1.92 participants per month. Twelve people did not meet the inclusion criteria, 11 declined, and one could not participate because of COVID-19-related issues.

Three participants were lost between the baseline and pre-surgery assessments because of public-health restrictions, leaving 11 in the intervention group and nine in usual care for those assessments. Recruitment began in October 2019, paused in March 2020, resumed in March 2021 and ended in September 2021. The original target was 50 patients, but only 23 were enrolled, below the minimum of 38 estimated for the registered non-inferiority analysis.

That calculation used a 2.0 ml·kg⁻¹·min⁻¹ margin, an assumed standard deviation of 1.8 ml·kg⁻¹·min⁻¹, 90% power and a one-sided significance level of 2.5%. Those who began the programme completed a median of five sessions, with a range of three to 11. The median interval from baseline assessment to surgery was 18.5 days, while enrolment to surgery took a median of 26.5 days, so the number of sessions was broadly consistent with the intended schedule of one every two to three days.

Total supervised exercise exposure was 63 hours. Five grade 1 adverse events occurred, equivalent to one event per 12.6 exercise hours. The researchers noted that this frequency was higher than expected and that comparisons with other studies are limited by differing definitions, surveillance practices and denominators.

Nevertheless, every event was mild: fatigue accounted for three reports and extremity pain for two. There were no grade 2 or more severe events, no serious adverse events, and no interruptions to exercise caused by adverse symptoms. The sample was too small to provide a precise estimate of uncommon harms, but within the participants studied, the supervised programme was tolerated without a serious safety signal.

Cardiorespiratory fitness, health-related quality of life and fatigue were assessed at baseline, immediately before surgery and 30 days after surgery. Fitness was measured with a symptom-limited cardiopulmonary exercise test on a cycle ergometer. The researchers determined peak oxygen uptake from the average of the highest consecutive 20-second values and identified the ventilatory threshold using a modified V-slope method confirmed through the ventilatory-equivalents approach.

Quality of life was measured with the Functional Assessment of Cancer Therapy–General questionnaire, covering physical, social and family, emotional and functional well-being, while the 13-item Functional Assessment of Chronic Illness Therapy–Fatigue scale quantified fatigue and its effects on daily activity. Mixed-effects models examined group-by-time changes, and Hedges’ g effect sizes with 95% confidence intervals were calculated for exercise-test outcomes. No significant group-by-time interaction was found for cardiorespiratory fitness, health-related quality of life or fatigue.

Given the small, heterogeneous sample and short intervention, these imprecise estimates neither demonstrate benefit nor establish that the programme has no effect. Post-operative results were likewise reported descriptively because the study was not powered to compare them. Within 30 days of surgery, six intervention participants and four usual-care participants experienced at least one complication.

Most were Clavien–Dindo grade I, meaning no treatment was required, although one grade III complication occurred. The median hospital stay was four days in both groups, and one 30-day readmission was recorded. The groups differed in several baseline characteristics, including habitual physical activity, body mass and the proportion receiving neoadjuvant chemotherapy, adding uncertainty to any between-group comparison.

Both groups also maintained physical well-being, fatigue and overall health-related quality of life from baseline to the pre-surgery assessment. The researchers could not separate an effect of exercise from other influences, including reassurance and support from repeated contact with an exercise professional, the burden of travelling to appointments, and possible disruption to daily routines. Standard quality-of-life measures may also incompletely capture concerns described by patients preparing for surgery, including preparedness, control and agency in recovery.

The recruitment data show that reaching patients in time was difficult at this regional centre. Although about half of the 47 people formally approached consented, only 47 of the 312 patients who underwent surgery with curative intent during the recruitment period were identified as potentially eligible and approached for screening. Patients cited travel burden, limited availability and short notice between enrolment and surgery as reasons for declining.

In a regional service, travel can require additional money, accommodation, time away from work or support from relatives and carers. Centre-based delivery enabled direct supervision and individual adjustment, but it also required dedicated staff, facilities and repeated journeys. The authors therefore propose that future studies examine broader referral pathways and remote or hybrid supervision, while preserving exercise monitoring and fidelity.

Cost-effectiveness and resource use were not evaluated, however, so the pilot cannot establish which delivery model is preferable or scalable. It also did not prespecify formal feasibility thresholds against which the programme could be judged to pass or fail, limiting objective interpretation of the recruitment and adherence results. The absence of a clear clinical signal should not be mistaken for evidence that prehabilitation is ineffective.

The exposure was brief: the median number of sessions was five before surgery, leaving limited time for exercise progression or physiological adaptation. The PROTECT sample also mixed breast, prostate and colorectal cancers, whose surgical pathways, perioperative care and expected complications differ, and was drawn from one centre. A single-centre design further limits transfer to services with different populations, staffing and referral systems.

Future work should therefore test specific cancer groups in adequately powered trials, examine strategies that produce a meaningful physiological stimulus within short surgical pathways, and evaluate remote or hybrid delivery as a possible response to geographic and time barriers.

Subject of Research: Safety and feasibility of supervised aerobic prehabilitation before cancer surgery

Article Title: Feasibility and safety of an exercise-based prehabilitation intervention before cancer surgery: a pilot randomised controlled trial

Article References: Hennessy, D. T., Shee, A. W., Rawstorn, J. C., Evans, L., Brown, S., & Wallen, M. P. (2026). Feasibility and safety of an exercise-based prehabilitation intervention before cancer surgery: a pilot randomised controlled trial. Supportive Care in Cancer, 34(11), Article 1132. https://doi.org/10.1007/s00520-026-11356-9

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11356-9

Keywords: prehabilitation, cancer surgery, supervised exercise, aerobic training, treatment safety, pilot randomised trial, cardiorespiratory fitness, postoperative outcomes, regional healthcare, exercise oncology, Feasibility, safety