Exercise as Medicine
Mechanistic justification and clinical application across androgen suppression and chemotherapy
Exercise is a biological intervention. It operates through identifiable molecular mechanisms — modifying the tumour microenvironment, defending muscle architecture against treatment-mediated catabolism, and directly counteracting the physiological consequences of androgen suppression and chemotherapy. These are not side benefits. They are primary effects.
Exercise is almost universally recommended to men undergoing treatment for prostate cancer. It appears in clinical guidelines, in oncology consultations, in patient information leaflets. The recommendation is correct. The justification is almost always inadequate.
The standard framing is welfare: exercise helps with fatigue, maintains weight, supports mood. This is true. But it locates exercise in the category of supportive care — something that makes treatment more tolerable — rather than in the category of biology. It treats exercise as comfort, not mechanism.
This paper argues for a different framing. Exercise is a biological intervention. It operates through identifiable molecular mechanisms. It modifies the tumour microenvironment, defends muscle architecture against treatment-mediated catabolism, and directly counteracts the physiological consequences of androgen suppression and chemotherapy. These are not side benefits. They are primary effects.
When exercise is understood as medicine, the question changes. The question is no longer whether to exercise, but how — what type, what intensity, what timing, and why those choices matter for the specific treatment context the individual is in.
Exercise is not a lifestyle recommendation appended to a medical protocol. It is a biological intervention with its own mechanistic justification — one that operates in parallel with, and sometimes in direct support of, pharmacological treatment.
Before addressing mechanisms, there is a prior question that most exercise oncology literature does not ask: why does exercise work at all? The answer is not that exercise is an enhancement layered onto normal biology. It is that human physiology evolved expecting intermittent physical stress. Movement is not an addition to the system — it is a signal around which the system was built.
The signalling cascades that exercise activates — AMPK, PGC-1α, myokine release, HPA axis calibration — are not incidental responses to an unusual stimulus. They are the organism's expected operating state. The sedentary condition, from this perspective, is not neutral. It represents the chronic absence of a required biological input — a form of signal deficiency that compounds over time in ways that conventional medicine does not categorise as disease but that nonetheless shift the terrain toward dysfunction. This framing is conceptual rather than directly evidenced at every step; its value lies in reorienting how the mechanistic evidence that follows is interpreted.
This framing matters for the prostate cancer context specifically. The QB framework treats cancer not as an intrusion into a previously healthy system, but as the downstream expression of a terrain that has lost its normal constraint mechanisms. Chronic physical inactivity removes one of the foundational constraint signals. Exercise restores it. In this sense, exercise is not a treatment for prostate cancer. It is a treatment for the biological terrain in which prostate cancer exists.
01Three mechanistic legs
The justification for exercise in the prostate cancer context rests on three distinct and complementary mechanistic arguments. They operate at different levels of biology and address different aspects of the disease and its treatment. None depends on the others. Together, they make the case for exercise as a foundational intervention that cannot be replaced by any pharmacological agent.
Leg 1: metabolic signalling — AMPK, mTOR, and insulin sensitivity
Physical exercise activates AMP-activated protein kinase (AMPK) — the cell's primary energy-sensing system — in proportion to the metabolic demand imposed. AMPK activation has several downstream consequences that are directly relevant to prostate cancer biology. [1,2]
AMPK phosphorylates and inhibits mTORC1, the central regulator of anabolic cell growth. In the context of a tumour microenvironment already characterised by constitutive PI3K/AKT/mTOR activation — driven in part by insulin excess, PTEN loss, and stromal signalling — exercise-mediated AMPK activation provides a periodic, physiological counter-signal. This is constraint through oscillation rather than chronic suppression: the signal rises during exercise and recovers in the rest period, preserving the cell's ability to respond to future signals rather than forcing permanent pathway inhibition. [3]
AMPK activation also improves peripheral insulin sensitivity, reducing the hyperinsulinaemia that drives AKT/MDM2-mediated p53 suppression — a central mechanism in the QB framework's account of prostate cancer terrain dysfunction. By lowering the chronic insulin load, exercise reduces one of the primary upstream drivers of the MDM2 nuclear stabilisation that suppresses p53 surveillance function. [4]
The arginine-polyamine axis: a fourth metabolic dimension
Preliminary preclinical evidence suggests a fourth metabolic dimension to Leg 1 that the field has not previously described. In prostate-specific Pten-knockout mice (Pb-Cre+; Pten(fl/fl)) — a model driven by the same PTEN-loss/AKT hyperactivation terrain that is central to the QB framework — voluntary exercise produced a 25% reduction in prostate weight and near-halving of Ki-67 proliferation (6.9% vs 16.5%, p=0.026), as reported by Oka et al. (AUA 2026, GG02-40). Body weight was identical between groups, ruling out caloric restriction as a confounding mechanism. [30]
RNA sequencing identified arginine catabolism pathways as the most significantly downregulated gene set in exercised mice. Arginase 1 (Arg1) — the enzyme catalysing the conversion of arginine to ornithine — was reduced at both mRNA and protein levels in tumour tissue. Metabolomic analysis confirmed suppression of the arginine-to-ornithine conversion in exercised animals. The downstream consequence is a substrate-level constraint on the polyamine synthesis cascade: arginine → ornithine (Arg1) → putrescine (ODC) → spermidine → spermine. The prostate is one of the highest polyamine-producing tissues in the body; polyamines are required for DNA replication, ribosome biogenesis, and cell proliferation. By suppressing Arg1, exercise starves this entire synthetic pipeline. [30]
The pharmacological rescue experiment provides internally compelling preclinical support. An arginase inhibitor alone replicated the exercise benefit (prostate weight 0.128g vs 0.161g, p=0.048; Ki-67 6.9% vs 13.9%, p=0.008). Adding exercise to arginase inhibition produced no further reduction (p=0.13 for weight; p=0.86 for Ki-67). In this model, exercise and arginase inhibition appear to operate on the same pathway — though this finding awaits replication in additional models and human tissue before it can be considered established. [30]
Preclinical support for exercise-mediated tumour suppression through the AMPK/metabolic axis is provided by voluntary wheel running studies in the TRAMP mouse model. Exercise produced a 60% reduction in high-grade tumour incidence compared to sedentary controls (71% of controls scored high-grade versus exercising mice), accompanied by significant reductions in circulating IL-1α, VEGF, and inflammatory cytokines. [5]
Leg 2: mitochondrial quality control
Exercise is the most potent physiological inducer of mitochondrial biogenesis and quality control. The mechanism runs through PGC-1α — the master transcriptional coactivator of mitochondrial gene expression — which is activated by both AMPK and p38 MAPK signalling downstream of muscle contraction. [6]
PGC-1α drives the production of new, healthy mitochondria while simultaneously upregulating mitophagy — the selective autophagic clearance of damaged mitochondria. This is the oscillatory stress-recovery cycle applied to organelle quality: exercise imposes metabolic demand that stresses mitochondria, recovery clears the damaged population, and repeated cycles progressively improve the functional fraction. [7]
This matters in the prostate cancer context for two reasons. First, tumour cells in a metabolically dysregulated microenvironment characteristically exhibit mitochondrial dysfunction — fragmented networks, reduced oxidative phosphorylation capacity, increased ROS production. The systemic effect of exercise-driven mitochondrial quality improvement in non-tumour tissue may contribute to a metabolic environment less supportive of tumour progression, though the direct mechanistic link between systemic mitochondrial improvements and tumour mitochondrial biology remains incompletely demonstrated. Second, in the setting of androgen deprivation and chemotherapy, where mitochondrial function in muscle is compromised by both hormonal changes and direct drug toxicity, exercise-mediated mitophagy and biogenesis provide a direct countermeasure to treatment-induced mitochondrial degradation. [6,8]
Leg 3: tumour microenvironment remodelling
The third mechanistic argument operates at the level of the tumour microenvironment. It encompasses three distinct and additive mechanisms at different levels of evidence maturity: the immunosuppressive myeloid axis (Arg1/arginine depletion), tumour vascular normalisation, and the cancer-associated fibroblast architectural axis (CNTF-mediated myofibroblast suppression).
3a. The Arg1-myeloid immunosuppressive axis (confirmed)
The Oka 2026 finding (described in Leg 1 above) carries a second, immunological dimension that operates independently of the polyamine proliferation story. Arginase 1 is not only a metabolic enzyme — it is a canonical immunosuppressive mediator. In the tumour microenvironment, Arg1 is expressed by tumour-associated macrophages (TAMs) in the M2 phenotype and by myeloid-derived suppressor cells (MDSCs). Arg1 activity in these cells depletes local arginine, and arginine is an amino acid that T cells and NK cells require for effector function: for TCR signalling, proliferation, and cytotoxic granule production. [30]
A TME depleted of arginine by Arg1-expressing myeloid cells is functionally immunosuppressed even in the absence of upregulated checkpoint ligands. Conversely, suppressing Arg1 in the TME restores arginine availability, re-enabling T cell and NK cell effector function. The strongest current hypothesis is that exercise reduces Arg1 in PCa tissue partly through reprogramming of the infiltrating myeloid compartment — shifting TAMs toward M1 phenotype and reducing MDSC representation — thereby contributing to restoration of immune competence in the tumour microenvironment. This remains to be confirmed by cell-type-specific analysis.
The cell-type specificity of the Arg1 reduction in Oka 2026 is not yet resolved — the study does not report immunofluorescence or cell-type deconvolution of the RNA-seq data. Whether the Arg1 suppression is tumour-cell-intrinsic, myeloid-mediated, or both is the most important open mechanistic question. Either answer is immunologically consequential.
3b. Tumour vascular normalisation (Strongly Supported)
An increasingly substantial body of evidence demonstrates that exercise normalises tumour vasculature — and this has direct consequences for drug delivery, immune cell access, and the hypoxic signalling that drives tumour aggression. Solid tumours characteristically develop dysfunctional, chaotic vasculature: vessels are irregularly dilated, poorly perfused, and leaky in ways that paradoxically impair rather than support adequate oxygenation. The result is tumour hypoxia — a condition that stabilises HIF-1α, promotes angiogenic signalling, selects for aggressive phenotypes, and creates a physical barrier to immune infiltration and cytotoxic drug delivery. [33,34]
Exercise — through shear stress on vessel walls, nitric oxide-mediated vasodilation, and VEGF pathway modulation — promotes vessel maturation and improved tumour perfusion. The shear stress mechanism has been characterised at the molecular level: aerobic exercise increases intravascular shear stress, activating calcineurin-NFAT-TSP1 signalling in tumour endothelial cells, which drives vessel remodelling and normalisation. Studies across multiple preclinical tumour models have shown that exercise increases tumour blood flow, reduces hypoxic fraction, and improves the spatial distribution of functional vessels. The functional consequences are threefold: reduced HIF-1α-driven aggressive signalling, improved penetration of cytotoxic agents into the tumour core, and improved access for immune effector cells. In mouse models, moderate aerobic exercise combined with chemotherapy produced significantly greater tumour growth inhibition than chemotherapy alone, attributable to improved drug delivery after vascular normalisation. [33,34,35]
3c. CAF architecture and CNTF — AACR 2026 abstract (Emerging)
Myofibroblastic CAFs are characterised by alpha-smooth muscle actin (α-SMA) expression and contractile activity. They are a primary driver of immune exclusion in solid tumours: they deposit dense extracellular matrix, compress vasculature, and create physical and signalling barriers that prevent CD8+ cytotoxic T cells from infiltrating the tumour core. The result is an immune-excluded, or cold, tumour microenvironment — one in which the adaptive immune system is present at the tumour periphery but cannot access its target. [9,10]
A novel and highly intriguing mechanism reported by Gong et al. (AACR 2026) suggests that exercise-induced CNTF signalling may suppress CAF myofibroblast transition. Using the TRAMP mouse model, Gong et al. report that exercise generates a myokine signal — ciliary neurotrophic factor (CNTF) — that suppresses the fibroblast-to-myofibroblast transition in cancer-associated fibroblasts. By maintaining CAFs in a quiescent rather than myofibroblastic state, exercise-induced CNTF may prevent the formation of α-SMA-expressing stromal barriers that exclude CD8+ T cells from the tumour core, potentially converting a cold TME toward a more immune-permissive state. Exogenous recombinant CNTF is reported to replicate this effect in sedentary animals and to synergise with immune checkpoint blockade. This finding awaits independent replication and full peer-reviewed publication; it is presented here as an emerging mechanism of considerable interest rather than established biology. [31]
The three mechanisms within Leg 3 are non-redundant and additive. The Arg1 pathway (3a) operates through myeloid metabolic reprogramming — restoring the arginine availability that T cells and NK cells require to function. The vascular normalisation pathway (3b) operates through vessel maturation and hypoxia reduction — improving immune and drug access to the tumour core. The CNTF pathway (3c) operates through fibroblast architectural reprogramming — suppressing the myofibroblast transition that creates structural barriers preventing immune cells from reaching their target. They address complementary failure modes of the immunosuppressed, physically excluded tumour microenvironment, at different levels of current evidence maturity.
Taken together, the three mechanistic legs address distinct biological levels: Leg 1 operates systemically through metabolic signalling and substrate-level proliferative constraints; Leg 2 operates at the organelle level through mitochondrial quality control; Leg 3 operates at the tumour microenvironment level through myeloid reprogramming, vascular normalisation, and CAF architectural remodelling. No currently established pharmacological intervention has been shown to address all three simultaneously.
Because the three legs draw on evidence of different maturities, it is worth being explicit about the confidence hierarchy. The following distinction runs throughout this paper and governs how mechanisms are described. Established mechanisms — AMPK activation, improved insulin sensitivity, resistance-training preservation of lean mass, fatigue reduction under ADT, cardiovascular protection — rest on decades of human RCT and mechanistic evidence and are described accordingly. Strongly supported mechanisms — mitochondrial quality control through PGC-1α/mitophagy, tumour vascular normalisation, myeloid TME modulation — are supported by substantial translational or animal evidence with meaningful human correlates. Emerging mechanisms — the Arg1/polyamine suppression axis and the CNTF/CAF reversion axis — are supported by 2026 preclinical conference abstracts with compelling internal validation; they are presented as the strongest current hypotheses in exercise-TME biology, not as settled science. A reader encountering a claim in this paper should be able to locate it within this hierarchy.
The three mechanistic legs are additive and non-redundant. But they share a structural feature that deserves explicit recognition: each one operates through oscillation rather than continuous activation. This is not incidental. It is the biological logic that makes exercise work — and it connects directly to one of the core principles of the QB framework.
Consider the sequence within a single exercise session. During exertion, AMPK spikes in proportion to energetic demand, suppressing mTORC1 and initiating catabolic processes. During recovery, mTOR is permitted to rise — driving protein synthesis, mitochondrial biogenesis, and adaptive remodelling in muscle. Mitophagy is activated in the recovery window: damaged mitochondria cleared during the stress phase are replaced by newly synthesised, higher-functioning organelles. The adaptation — increased mitochondrial density, improved insulin sensitivity, stronger muscle fibres — occurs not during the stress but during the recovery that follows it.
The body does not improve under chronic stress, nor under chronic rest. It improves under stress followed by recovery. Exercise is the deliberate imposition of the stress signal. Recovery is the condition under which adaptation occurs. Neither works without the other.
This is precisely the oscillation principle that the QB framework applies across its protocol design — from rapamycin dosing to doxycycline cycling to testosterone restoration. Exercise is not a special case of this principle. It is its clearest natural expression. The sedentary state fails not because it involves too much stress but because it removes the oscillatory signal entirely, leaving biology in a condition of chronic under-stimulation from which adaptive improvement cannot occur.
The clinical implication is that exercise timing and recovery design are not peripheral considerations — they are mechanistically central. The prescription is not simply "exercise more." It is: impose the right stress signal, then protect the recovery window in which the adaptation is built.
Each of the mechanistic legs described above has a pharmacological partial analogue. The table below summarises the coverage — and the gaps.
No single agent addresses all three mechanistic legs simultaneously. No currently available pharmacological strategy has been shown to reproduce the full integrated biological response generated by exercise — a whole-system signal that activates metabolic, mitochondrial, and immune-architectural remodelling in a single coordinated response.
The pioglitazone row in the table above deserves specific note. As discussed in Leg 3, pioglitazone converges with exercise on the myeloid Arg1 node through a different upstream mechanism (PPARγ/NF-κB rather than exercise-mediated myokine signalling). This is complementarity, not redundancy: the two interventions approach the same target from different angles, and there is no pharmacological rationale for treating their convergence as grounds to reduce either.
Exercise is not one mechanism. It is a coordinated biological event — one that evolution built into the organism's expected operating conditions and that no pharmacological agent yet replicates in full. This is not an argument against pharmacological support. Metformin, rapamycin, and GLP-1 agonists each have their role in the QB protocol — and their mechanisms are complementary to exercise rather than competitive with it. The argument is that exercise cannot be omitted on the assumption that drugs cover the same ground. They do not. They cover adjacent ground.
Most oncology trials ask a single question: does this drug improve survival? INTERVAL-GAP4 asked a structurally different question: can structured, supervised exercise — delivered as a defined intervention in men with metastatic castration-resistant prostate cancer — improve overall survival? The trial was multicentre, randomised, and phase III, with overall survival as its primary endpoint — not quality of life or fatigue, the endpoints to which exercise has historically been confined in oncology trials. INTERVAL-GAP4 closed to accrual in February 2023 having enrolled 145 participants against a target of 866, due to recruitment and operational challenges amplified by COVID-19. Survival follow-up is ongoing. [32]
If INTERVAL-GAP4 demonstrates a survival benefit in its follow-up analysis, the implications extend beyond prostate cancer. It would establish exercise as a survival-modifying intervention in a metastatic solid tumour — a standard of evidence that most pharmacological agents in this space do not meet. The underpowered enrolment will limit statistical confidence, but the biological question the trial was designed to ask remains valid and the follow-up data will be informative.
INTERVAL-GAP4 was not testing whether exercise makes cancer treatment more tolerable. It was testing whether exercise changes the biology of cancer in a way that extends life. That is a different question from anything the exercise oncology literature had previously asked at this scale. The answer, when the follow-up data mature, will either confirm the mechanistic argument or qualify it in ways that refine the prescription. Either outcome is scientifically useful.
The trial deserves attention not because its result is known — survival follow-up is ongoing — but because the question it was designed to ask represents a genuine shift in how exercise is positioned within oncology. [32]
02Exercise under androgen deprivation therapy
Androgen deprivation therapy — whether achieved through GnRH agonists, GnRH antagonists, or bilateral orchiectomy — produces a rapid and profound shift in systemic physiology that extends far beyond the intended reduction in testosterone. The felt consequences are well known to anyone who has experienced them or spoken at length to someone who has: fatigue, loss of lean muscle mass, weight gain, cognitive change, hot flushes, mood disturbance, loss of libido and erectile function, bone density reduction. These are not minor inconveniences. They are significant alterations in the experience of being in a body.
Exercise does not eliminate these consequences. But it acts on their biology in ways that are specific, measurable, and clinically meaningful.
Fatigue
Cancer-related fatigue under ADT has a specific biological signature: elevated inflammatory cytokines (IL-6, TNF-α, IL-1β), disrupted hypothalamic-pituitary-adrenal axis function, and reduced mitochondrial energy production capacity in skeletal muscle. Exercise addresses all three. Aerobic training reduces circulating inflammatory cytokines, restores HPA axis sensitivity, and — through PGC-1α/mitochondrial biogenesis — improves the muscle's capacity to generate ATP aerobically rather than relying on less efficient anaerobic pathways. [11]
RCT evidence is robust. The ERASE trial demonstrated that high-intensity interval training significantly reduced cancer-related fatigue in men on ADT compared to moderate continuous exercise, with effects sustained at follow-up. The dose-response relationship between exercise intensity and fatigue reduction is now well-established: higher intensity produces larger effects, within the limits of tolerance. [12]
Muscle loss and sarcopenia
Testosterone is a primary anabolic signal in skeletal muscle — it promotes protein synthesis through androgen receptor-mediated gene expression, suppresses myostatin (a negative regulator of muscle growth), and supports satellite cell activation for muscle repair. ADT removes this signal. The result, without countermeasure, is progressive loss of lean muscle mass — sarcopenia — that compounds over the duration of treatment. [13]
Resistance training is the most effective countermeasure available. It activates muscle protein synthesis through mechanical load-mediated mTORC1 activation in muscle — independent of androgenic signalling. It suppresses myostatin through IGF-1 and follistatin upregulation. It recruits and activates satellite cells through mechanical and paracrine signals. In men on ADT, resistance training consistently preserves or partially restores lean muscle mass in RCTs, with effect sizes that are clinically meaningful. [14]
The timing and composition of protein intake relative to exercise is an important adjunct. Post-exercise protein synthesis is maximal in the 3–4 hours following resistance exercise; adequate protein provision — particularly leucine-rich sources — during this window supports the anabolic stimulus. This is relevant to protocol design: exercise timing should be coordinated with nutritional strategy rather than treated as isolated. [15]
Bone density
ADT accelerates bone mineral density loss through reduced osteoblast activity and relatively increased osteoclast activity — a consequence of the loss of testosterone's indirect (oestrogen-mediated) protective effect on bone. Weight-bearing and resistance exercise provides mechanical loading that stimulates osteoblast activity and bone remodelling, partially offsetting ADT-induced bone loss. The effect is site-specific: loading must be applied to the relevant skeletal regions. [16]
Cardiovascular and metabolic risk
ADT induces a cluster of metabolic changes — increased adiposity, reduced insulin sensitivity, dyslipidaemia, increased cardiovascular risk — that overlap substantially with metabolic syndrome. Aerobic exercise directly addresses each component: it improves insulin sensitivity, reduces visceral adiposity, improves lipid profiles, and reduces resting cardiovascular risk markers. The effect of ADT on cardiovascular mortality is a recognised clinical concern; exercise is the most evidence-based mitigation available. [17]
Mood, cognition, and vasomotor symptoms
ADT-associated mood disturbance, cognitive change, and hot flushes share a common upstream cause: the withdrawal of testosterone and its aromatised derivative oestradiol from central nervous system function. Exercise partially compensates through multiple routes — BDNF upregulation supports cognitive function and mood, endorphin and endocannabinoid release reduces anxiety and improves sleep architecture, and thermoregulatory adaptations from aerobic training reduce the severity of vasomotor events. These effects are real but more variable across individuals than the muscle and fatigue effects. [18]
03Exercise under androgen receptor pathway inhibition
Enzalutamide, abiraterone, darolutamide, and apalutamide extend androgen suppression beyond the GnRH axis — blocking androgen receptor function directly (enzalutamide, darolutamide, apalutamide) or eliminating residual androgen synthesis in adrenal tissue and tumour cells themselves (abiraterone). They compound the physiological consequences of ADT rather than replacing them.
Enzalutamide in particular has a distinctive central nervous system profile that adds to the exercise challenge: it crosses the blood-brain barrier and can cause fatigue, cognitive slowing, and — rarely but significantly — seizure risk. The fatigue under enzalutamide is qualitatively different from ADT fatigue alone, and is one of the primary reasons for treatment discontinuation. High-intensity exercise has a complex interaction with CNS fatigue: it is likely beneficial at moderate doses but requires individual calibration, and the ERASE-type intensity prescription may need modification for men on enzalutamide specifically. [12,19]
Abiraterone's primary additional consideration is mineralocorticoid excess — it requires co-administration of prednisone or methylprednisolone, which themselves carry catabolic and metabolic effects on muscle. This further amplifies the sarcopenia risk and underscores the importance of resistance training in this treatment context. [20]
The exercise prescription principles from Section 2 apply in full under ARPI — resistance training for muscle preservation, aerobic training for fatigue and cardiovascular protection — with the addition of careful monitoring of CNS-related symptoms and individual tolerance calibration.
04Exercise under chemotherapy
Docetaxel and cabazitaxel — the two taxane agents used in castration-resistant prostate cancer — produce a distinct and in some respects more acute physiological challenge than hormonal suppression. Their mechanism (microtubule stabilisation, preventing mitotic spindle disassembly) is non-selective: it affects all rapidly dividing cells, including those in bone marrow, gut epithelium, and peripheral nerve tissue. The practical consequences are neutropenia, gastrointestinal toxicity, peripheral neuropathy, and pronounced fatigue.
Exercise under chemotherapy requires a different risk-benefit calculation than exercise under hormonal suppression. It is not contraindicated — the evidence increasingly supports its benefit — but it must be timed, dosed, and structured with treatment cycles in mind.
Timing within treatment cycles
Chemotherapy-induced neutropenia typically reaches its nadir approximately 7–14 days after infusion, depending on regimen and individual factors — docetaxel tends to nadir earlier, around day 7. [36] During this window, high-intensity exercise carries infection risk and should be modified or replaced with low-intensity movement. In the week before infusion and from day 14 onward (as counts recover), higher intensity exercise is appropriate and beneficial. Exercise during chemotherapy remains beneficial but should be modified according to treatment-related toxicities, including periods of significant immunosuppression. [21]
Peripheral neuropathy
Taxane-induced peripheral neuropathy (TIPN) affects balance, proprioception, and the ability to perform lower-limb weight-bearing exercise safely. Balance training becomes relevant as an addition to the standard prescription — not merely for the therapeutic effect on neuropathy (which is limited) but to reduce fall risk in a population already at elevated risk from bone density loss and muscle weakness. Exercise in water (aquatic resistance training) is an effective adaptation for men with significant TIPN, removing ground reaction forces while preserving resistance stimulus. [22]
Muscle wasting under chemotherapy
Docetaxel and cabazitaxel both induce systemic inflammation — elevated IL-6, TNF-α, and other catabolic signals — that drives muscle protein breakdown through the ubiquitin-proteasome pathway (specifically MuRF-1 and MAFbx/atrogin-1 E3 ligases). This is the same pathway implicated in cancer cachexia more broadly. Resistance training suppresses these ligases through IGF-1/PI3K/AKT/FoxO3a signalling — FoxO3a transcription factor activation is required for MuRF-1 and atrogin-1 expression, and AKT phosphorylation of FoxO3a sequesters it in the cytoplasm, preventing this transcription. [23]
Myostatin, the primary negative regulator of skeletal muscle mass, is elevated under chemotherapy-induced inflammatory conditions. Resistance exercise suppresses myostatin and upregulates its antagonist follistatin. In the TRAMP model, exercising mice maintained significantly greater lean muscle mass and grip strength than sedentary controls across a 20-week treatment period — with myostatin significantly lower in the exercise group. [24]
The practical prescription under chemotherapy therefore prioritises resistance training for muscle preservation — modified in intensity to cycle timing and tolerance — with aerobic training for fatigue and cardiovascular support in the non-nadir windows.
Fatigue under chemotherapy
Chemotherapy-induced fatigue has a different biology from ADT fatigue: it is more acute, more variable across cycles, and more closely tied to inflammatory cytokine peaks. The counterintuitive finding — consistent across multiple RCTs — is that exercise during chemotherapy reduces fatigue rather than worsening it, despite the intuitive assumption that rest is needed. The mechanism is likely a combination of anti-inflammatory cytokine effects, improved sleep quality, and maintenance of cardiovascular fitness that prevents the deconditioning spiral. [25]
The case for exercise under chemotherapy is not that it makes a difficult experience easier, though it does. It is that it preserves the physiological substrate — muscle architecture, mitochondrial function, cardiovascular capacity — on which the next phase of treatment and recovery depends.
05Toward an exercise prescription
The evidence does not support a single universal exercise prescription across treatment contexts. What it supports is a principle-based approach in which exercise type, intensity, and timing are selected to match the specific physiological challenge of the treatment phase.
Individual calibration based on baseline fitness, treatment tolerance, and clinical context is essential. The table above represents a framework for that calibration — not a fixed protocol. The prescription should be treated as a starting point that is adjusted iteratively based on response, with recovery quality (sleep, muscle soreness, fatigue trajectory) used as the primary feedback signal.
06Protein requirements and the mTOR objection
The exercise prescription described in Sections 2 to 5 produces its muscle-preserving effect only when adequate protein substrate is available. In the prostate cancer context this is not straightforward, because a version of the protein-promotes-cancer argument is specifically prevalent in this patient population and directly implicates this paper's central mechanistic framework.
The argument runs as follows: dietary protein raises circulating IGF-1; IGF-1 activates PI3K/AKT/mTOR through the insulin receptor and IGF-1R; mTOR drives anabolic cell growth; therefore protein intake promotes prostate cancer progression through the same AKT/mTOR axis described throughout this paper. Each step has biological support. The clinical conclusion does not follow.
The acute/chronic distinction
The constitutive mTOR activation that characterises prostate tumour biology arises from chronic AKT signalling driven by PTEN loss, insulin resistance, and hyperinsulinaemia. This is a persistent, unregulated state that operates independently of meal timing, and it is addressed in the QB framework by insulin sensitivity improvement and AMPK-mediated mTORC1 counter-signalling.
The mTOR activation that follows a protein-containing meal is acute, transient, and oscillatory. Leucine activates mTORC1 in skeletal muscle through the Ragulator/mTORC1 amino acid sensing pathway, a response that peaks and resolves in the post-prandial window, driving muscle protein synthesis before returning to baseline. This is the normal physiological mechanism through which skeletal muscle maintains its mass. It is not the same biology as constitutive oncogenic mTOR activation driven by upstream AKT dysregulation. Treating the two as equivalent is a categorical error with measurable clinical consequences. [4,27]
The epidemiological evidence
The primary epidemiological source cited in support of protein restriction in cancer patients is Levine et al. (Cell Metabolism, 2014), which found that high animal protein intake was associated with elevated cancer mortality in adults aged 50 to 65. The finding is real. What is almost never cited from the same paper is the age reversal: in adults aged 65 and above, high protein intake was associated with reduced cancer mortality and reduced overall mortality. The paper that is most often used to argue against protein in prostate cancer patients simultaneously argues for it in the relevant population. [26]
A 2024 umbrella review of systematic reviews conducted for the German Nutrition Society guidelines found no association between total protein intake and cancer incidence across breast, prostate, colorectal, ovarian, and pancreatic cancers. Neither animal nor plant protein intake was associated with increased cancer risk at any evidence grade. [28]
Practical targets under ADT, ARPI, and chemotherapy
In men on ADT, sarcopenia develops progressively without resistance training and adequate protein. Under abiraterone with mandatory prednisone co-administration, catabolic pressure is further amplified. Under taxane chemotherapy, inflammatory signalling elevates ubiquitin-proteasome pathway activity. In all three contexts, protein requirements are higher than normal because endogenous anabolic signalling is suppressed.
The clinical target is 1.6g protein per kilogram of body weight daily as a minimum, rising to 1.8–2.0g/kg under active catabolic treatment. These targets are consistent with ESPEN oncology guidelines and are supported by evidence that intakes below 1.2–1.4g/kg are associated with ongoing muscle wasting during cancer treatment even when formally within standard dietary recommendations. [15,29]
Protein source matters less than total amount. Unprocessed meat, fish, eggs, dairy, legumes, and protein supplements all contribute to the target. Protein distribution across the day is more effective than concentration in one meal; leucine-rich sources in the 3–4 hours post-resistance training maximise the anabolic window that exercise creates. [15]
07The Quiet Biology position
Exercise occupies a foundational position in the QB framework — not as an adjunct to the pharmacological protocol, but as a primary intervention that addresses biological terrain in ways that pharmacology cannot replicate.
The evolutionary argument establishes why: exercise is not an enhancement but a required biological input. The sedentary state is not neutral. It is the chronic absence of a signal around which human physiology organised itself. Restoring that signal restores terrain. And terrain is what the QB framework treats.
The three mechanistic legs are each independently justified, operating at different levels of evidence maturity. Leg 1 — AMPK activation, insulin sensitivity improvement, and mTORC1 oscillation — rests on decades of established mechanistic and translational evidence; preliminary preclinical evidence from the Pten-KO model (Oka et al., AUA 2026) adds a potentially important Arg1/polyamine dimension that awaits replication. Leg 2 — mitochondrial quality control through PGC-1α and exercise-induced mitophagy — is strongly supported by human and animal data as a mechanism of treatment-induced mitochondrial defence. Leg 3 draws on two 2026 preclinical conference abstracts: the Arg1/myeloid TME hypothesis (Oka et al.) and the CNTF/CAF reversion hypothesis (Gong et al., AACR 2026) — both biologically compelling and pharmacologically validated at the preclinical level, both awaiting full publication and independent replication. These are presented as the strongest current emerging hypotheses in exercise-TME biology, not as settled mechanisms.
No pharmacological agent addresses all three legs simultaneously. Metformin activates AMPK but cannot replicate mechanical loading. Rapamycin suppresses mTOR but does not drive mitochondrial biogenesis. GLP-1 agonists improve insulin sensitivity but do not independently preserve muscle. Myostatin inhibitors preserve muscle but do not improve cardiovascular or mitochondrial function. Pioglitazone converges with exercise on the Arg1/myeloid TME axis — an important complementarity — but does not address Legs 1 or 2. Exercise covers all of this ground at once, and it does so through the oscillatory stress-recovery architecture that is the biological logic of adaptation itself.
Protein is the substrate on which that adaptation is built. The mTOR objection — that post-exercise protein drives the same pathway implicated in tumour growth — conflates two biologically distinct events: the acute, self-resolving mTORC1 activation of muscle protein synthesis, and the constitutive, upstream-driven mTOR dysregulation of tumour biology. They share a pathway name. They are not the same phenomenon.
- Hardie DG. AMPK: a target for drugs and natural products with effects on both diabetes and cancer. Diabetes. 2013;62(7):2164–2172.
- Steinberg GR, Carling D. AMP-activated protein kinase: the current landscape for drug development. Nat Rev Drug Discov. 2019;18(7):527–551.
- Faubert B, et al. AMPK is a negative regulator of the Warburg effect and suppresses tumor growth in vivo. Cell Metab. 2013;17(1):113–124.
- Pollak M. Insulin and insulin-like growth factor signalling in neoplasia. Nat Rev Cancer. 2008;8(12):915–928.
- Meyers JA, et al. Nexrutine and exercise similarly prevent high grade prostate tumors in transgenic mouse model. PLoS One. 2019;14(12):e0226187. PMID 31856170.
- Laye MJ, et al. Exercise and mitochondrial health. J Physiol. 2019;597(20):5139–5150.
- Drake JC, et al. Mitochondria quality control in the skeletal muscle: the role of exercise. Exerc Sport Sci Rev. 2016;44(3):107–116.
- Hojman P, et al. Exercise effects on cancer — biochemical and cellular mechanisms. Nat Rev Cancer. 2018;18(12):669–680.
- Sahai E, et al. A framework for advancing our understanding of cancer-associated fibroblasts. Nat Rev Cancer. 2020;20(3):174–186.
- Chakravarthy A, et al. Pan-cancer deconvolution of tumour composition using DNA methylation. Nat Commun. 2018;9(1):3220.
- Bower JE. Cancer-related fatigue: mechanisms, risk factors, and treatments. Nat Rev Clin Oncol. 2014;11(10):597–609.
- Fairman CM, et al. (ERASE trial). High-intensity exercise and cancer-related fatigue in men with prostate cancer on androgen deprivation therapy. JAMA Oncol. 2021;7(8):1150–1157.
- Grossmann M, et al. Muscle and metabolic effects of androgen deprivation therapy in prostate cancer. Best Pract Res Clin Endocrinol Metab. 2013;27(4):549–564.
- Galvao DA, et al. Combined resistance and aerobic exercise program reverses muscle loss in men undergoing androgen suppression therapy for prostate cancer without bone metastases. J Clin Oncol. 2010;28(2):340–347.
- Moore DR, et al. Ingested protein dose response of muscle and albumin protein synthesis after resistance exercise in young men. Am J Clin Nutr. 2009;89(1):161–168.
- Galvao DA, et al. Resistance training and reduction of treatment side effects in prostate cancer patients. Med Sci Sports Exerc. 2006;38(12):2045–2052.
- Keating NL, et al. Diabetes and cardiovascular disease during androgen deprivation therapy for prostate cancer. J Clin Oncol. 2006;24(27):4448–4456.
- Segal RJ, et al. Resistance exercise in men receiving androgen deprivation therapy for prostate cancer. J Clin Oncol. 2003;21(9):1653–1659.
- Shore ND, et al. Fatigue with enzalutamide: mechanism and clinical management. Prostate Cancer Prostatic Dis. 2018;21(3):311–321.
- Ryan CJ, et al. (COU-AA-302). Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J Med. 2013;368(2):138–148.
- Campbell KL, Winters-Stone KM, Wiskemann J, et al. Exercise Guidelines for Cancer Survivors: Consensus Statement from International Multidisciplinary Roundtable. Med Sci Sports Exerc. 2019;51(11):2375–2390. DOI: 10.1249/MSS.0000000000002116.
- Streckmann F, et al. Exercise program improves therapy-related side effects and quality of life in lymphoma patients undergoing therapy. Ann Oncol. 2014;25(2):493–499.
- Bodine SC, et al. Identification of ubiquitin ligases required for skeletal muscle atrophy. Science. 2001;294(5547):1704–1708.
- Thomas RJ, et al. Exercise preserves muscle mass and force in a prostate cancer mouse model. J Cachexia Sarcopenia Muscle. 2019;10(5):1097–1107.
- Courneya KS, et al. Effects of aerobic and resistance exercise in breast cancer patients receiving adjuvant chemotherapy. J Clin Oncol. 2007;25(28):4396–4404.
- Levine ME, et al. Low protein intake is associated with a major reduction in IGF-1, cancer, and overall mortality in the 65 and younger but not older population. Cell Metab. 2014;19(3):407–417.
- Zoncu R, et al. mTOR: from growth signal integration to cancer, diabetes and ageing. Nat Rev Mol Cell Biol. 2011;12(1):21–35.
- Kühn T, et al. Protein intake and cancer: an umbrella review of systematic reviews for the evidence-based guideline of the German Nutrition Society. Eur J Nutr. 2024;63(5):1471–1486.
- Arends J, et al. ESPEN guidelines on nutrition in cancer patients. Clin Nutr. 2017;36(1):11–48.
- Oka T, Hatano K, Hase H, Konishi M, et al. Exercise inhibits prostate cancer progression by suppressing Arginase 1. GG02-40. Journal of Urology, AUA Annual Meeting Supplement. 2026. DOI: 10.1097/01.JU.0001191424.49345.3c.40.
- Gong P, Liu Y, Han L, Yang L, Lin C, et al. CNTF governs exercise-induced myofibroblast reversion in prostate cancer. Cancer Research. 2026;86(7 Suppl):Abstract nr 2098. Presented at: AACR Annual Meeting 2026.
- Newton RU, Hart NH, Galvao DA, et al. Intense exercise for survival among men with metastatic castrate-resistant prostate cancer (INTERVAL-GAP4): a multicentre, randomised, controlled phase III study protocol. BMJ Open. 2018;8(5):e022899. DOI: 10.1136/bmjopen-2018-022899. NCT02730338. Trial closed to accrual February 2023; n=145 enrolled; survival follow-up ongoing. Feasibility report: Newton RU et al. Eur Urol Open Sci. 2026.
- Schadler KL, Thomas NJ, Galie PA, et al. Tumor vessel normalization after aerobic exercise enhances chemotherapeutic efficacy. Oncotarget. 2016;7(40):65429–65440. DOI: 10.18632/oncotarget.11748.
- Delaney G, et al. Exercise modulation of tumour perfusion and hypoxia to improve radiotherapy response in prostate cancer. Prostate Cancer Prostatic Dis. 2021;24(2):318–328. DOI: 10.1038/s41391-020-0245-z.
- Rigby M, et al. Vascular normalisation as the stepping stone into tumour microenvironment transformation. Br J Cancer. 2021;125(8):1106–1116. DOI: 10.1038/s41416-021-01330-z.
- Crawford J, Dale DC, Lyman GH. Chemotherapy-induced neutropenia: risks, consequences, and new directions for its management. Cancer. 2004;100(2):228–237. DOI: 10.1002/cncr.11882.
- Patient-facing companion to this paper. Plain language guide for men on hormone suppression or chemotherapy.
- ActiveRead phase →
- RefractoryRead phase →