Preoperative Exercise Training Did Not Increase Surgery Rates in 207-Patient Trial

A Danish randomized trial asked whether preoperative exercise during chemotherapy changes who actually reaches curative cancer surgery. Across 207 patients, preoperative exercise did not shift those odds at all.

Preoperative Exercise is one of the most intuitive ideas in modern cancer care: make a patient fitter while chemotherapy is running, and more of them should be well enough to undergo the operation that offers a cure. A randomized clinical trial published in JAMA Network Open on October 28, 2026, tested that logic head-on in gastroesophageal cancer, and the result came back flat.

The trial was run from Copenhagen University Hospital, Rigshospitalet. It randomized 207 patients with gastroesophageal adenocarcinoma 1:1 to preoperative exercise training plus usual care, or to usual care alone. Of those, 202 were analyzed. Enrollment ran from August 16, 2018, to September 23, 2022, with final follow-up on October 22, 2024, and the trial is registered as NCT03490565.

The primary question was not fitness or strength. It was whether patients made it to the operating table at all, measured as failure to undergo an intended curative tumor resection. On that outcome, the odds ratio was 1.28 (95% CI, 0.48-3.38; P = .62) – numerically in the unfavorable direction, and nowhere near statistical significance.

The investigating group was led by Casper Simonsen, with Jesper Frank Christensen as senior author. It drew on the Centre for Physical Activity Research alongside the departments of digestive diseases, oncology, pathology and radiology at Rigshospitalet, and the Danish Cancer Institute.

What the Preoperative Exercise Program Actually Involved

This was not a pamphlet-and-encouragement intervention. The preoperative exercise consisted of two weekly sessions of supervised concurrent resistance training and aerobic exercise, delivered by physiotherapists or exercise physiologists either at the research center or at municipality-based health care centers.

The aerobic half was high-intensity interval training on a stationary bicycle. After a 10-minute warm-up, participants completed three four-minute intervals above 85% of maximal heart rate, separated by three-minute active rests. That is a demanding prescription for anyone, let alone someone mid-chemotherapy.

The resistance half started conservatively, with two sets of eight repetitions of step-ups and squats against light resistance. It then progressed to three sets of eight to 12 repetitions at 50% to 80% of one-repetition maximum on leg press, leg extension, chest press and seated row.

The Outcome That Refused to Move

Ten of 101 participants in the preoperative exercise arm (9.9%) did not undergo the intended curative tumor resection. The comparison group fared marginally better, which is why the odds ratio sits above 1.00 rather than below it, but the confidence interval spans a range so wide that it accommodates both meaningful harm and meaningful benefit.

Secondary outcomes told the same story. Postoperative complications showed no between-group difference (OR, 1.16; 95% CI, 0.61-2.20). Neither did permanent discontinuation of neoadjuvant treatment (OR, 0.81; 95% CI, 0.38-1.72), the outcome where a fitter patient might most plausibly have gained ground.

Why Recruitment Stopped Short of Target

The trial was designed around a specific statistical requirement: 147 participants in each group were needed to achieve 80% power in a two-sided test at an alpha of .05. It never reached that. The principal investigator terminated inclusion in October 2022 because funding had run out, leaving the study at 66.8% of its target sample.

That matters for how the null result should be read. An underpowered trial that finds nothing has not demonstrated that nothing is there; it has failed to resolve the question at the precision it set out to achieve.

Attendance Was Uneven Across the Preoperative Exercise Arm

Participants attended a median of 14 sessions (IQR, 8-19), with a median attendance rate of 76.1% (IQR, 45.7%-90.0%). Those interquartile ranges are the detail worth pausing on: a quarter of participants attended fewer than 45.7% of their scheduled sessions.

Seven participants (6.9%) never received the intervention at all, and 21 (20.6%) permanently discontinued it. A trial of preoperative exercise under these conditions is testing an offer of training as much as training itself, which is a reasonable thing to test but a different question from whether the physiology works.

What a Null Primary Outcome Does and Does Not Rule Out

The authors concluded that preoperative exercise training did not increase the likelihood of undergoing tumor surgery. That is a precise claim about one endpoint in one cancer population, and it should not be stretched.

It does not say that exercise during cancer treatment is useless. Reaching surgery is a blunt, binary outcome governed largely by disease progression and treatment toxicity – factors that twice-weekly training may simply not touch. Our earlier coverage of structured exercise and cancer survival across pooled studies looked at a very different endpoint over a much longer horizon.

Nor does it bear on quality of life, strength or fitness during treatment, none of which was the primary outcome here. A separate trial we covered on exercise and stamina in prostate cancer addressed exactly those patient-facing measures.

How the Program Compares With Other Interval and Resistance Protocols

The bicycle protocol used here is recognizable from mainstream interval research: four-minute efforts above 85% of maximal heart rate are a standard dose in studies of interval training and blood vessel function. The resistance prescription, at 50% to 80% of one-repetition maximum, likewise sits within conventional ranges used in resistance training work with older adults.

In other words, the intervention was not exotic or underdosed on paper. What it faced was a population undergoing chemotherapy for an aggressive cancer, and an endpoint that may lie largely outside the reach of training.

The Surgical Context Behind the Endpoint

Gastroesophageal adenocarcinoma is treated with chemotherapy first and surgery second, and the gap between those two stages is where patients are lost. Disease that progresses during chemotherapy, toxicity that will not clear, or a decline in overall condition can all remove the operation from the table.

That is the pathway a preoperative exercise program is implicitly trying to interrupt. The logic is that greater cardiorespiratory reserve and preserved muscle make a patient a better surgical candidate, and make them more likely to tolerate the full chemotherapy course. This trial measured both links in that chain and found neither one moved.

It is worth noting what the design got right. The outcome was hard, objective and clinically meaningful rather than a surrogate measure, the comparison was usual care rather than a sham, and the follow-up extended to October 2024. Trials of preoperative exercise have often leaned on fitness markers instead; this one went after the decision that actually changes a patient outcome.

What This Means for Anyone Weighing Preoperative Exercise

For readers facing major surgery, the practical reading of this trial is narrow and fairly undramatic. It gives no reason to expect that training before an operation will determine whether that operation goes ahead, and it also gives no reason to avoid reasonable activity.

• Decisions about exercise during cancer treatment belong with the treating oncology and surgical team, who know the individual disease picture.
• A null result on one binary endpoint is not evidence of harm; postoperative complications also showed no between-group difference.
• Adherence in this trial was genuinely difficult for many participants, which is itself useful information when planning anything during chemotherapy.
• Goals other than reaching surgery – maintaining strength, function or independence – were not what this trial measured.

The broader field of preoperative exercise research is still young, and a trial that stopped at two-thirds of its intended size is a contribution to that field rather than a verdict on it.