Upright vs Supine Particle Therapy: Clinical Indications and Configuration Implications

• Upright positioning achieves comparable setup accuracy to supine for head and neck targets, with early clinical outcomes in re-irradiation and paediatric craniospinal irradiation demonstrating feasibility and significant dosimetric advantages over photon plans.
• Anatomical changes in the upright posture increase lung volume by 5–15% and reduce breathing motion amplitude, but require site-specific immobilisation and beam angle optimisation to realise benefits for thoracic and abdominal treatments.
• The current evidence supports upright delivery for selected cranial, head and neck, and potentially thoracic indications, but gantry-based systems remain necessary for complex multi-field cases; a hybrid configuration of one gantry room and multiple upright rooms may optimise cost and clinical coverage.

The debate between upright and supine patient positioning in particle therapy has shifted from a historical footnote to a central question in facility design. As proton and carbon ion centres proliferate, the rotating gantry — a multi-million-dollar, multi-storey mechanical structure — has become the dominant cost driver and spatial constraint. Upright, chair-based delivery using a fixed horizontal beamline offers a radically simpler alternative. But simplicity is not the same as equivalence. A growing body of evidence now allows us to compare the two postures across anatomical sites, immobilisation requirements, dosimetric outcomes, and clinical feasibility. This article synthesises that evidence to identify where upright treatment can replace the supine standard, where it cannot, and what the implications are for facility configuration.

Anatomical and Technical Considerations

The transition from supine to upright posture produces measurable, site-specific changes in internal anatomy that carry both opportunities and challenges. Volz et al. (2024) reviewed the literature across photon and particle therapy, finding that lung volume increases by 5.3–14.7% in the upright position, with reduced diaphragm excursion and breathing motion amplitude. The liver assumes a more inferior position, and the prostate shows heightened stability, minimally influenced by bladder filling. For lung cancer specifically, Martire et al. (2026) reported a 13% increase in total lung volume and halving of target motion in one of six thoracic patients when comparing paired upright and supine 4DCTs. These anatomical shifts are not trivial — they directly affect target coverage and organ-at-risk sparing.

Immobilisation is the critical technical hurdle. Volz et al. (2022) noted that conventional supine devices such as vacuum bags and thermoplastic masks cannot be directly transferred to the upright posture; vacuum bag contents settle unevenly when vertical, and masks may not provide adequate weight-bearing support. Varnava et al. (2025) evaluated three immobilisation setups for upright carbon ion therapy in ten healthy volunteers. The combination of vacuum bags and a thermoplastic shell reduced interfractional Euclidean distances to 5.5 mm (abdominal) and 2.9 mm (head and neck), compared to over 11 mm with no immobilisation. Intrafractional motion over 15 minutes was similarly improved, with mean shifts of 2.1–2.2 mm for the shell setup. Comfort ratings remained acceptable across all setups, indicating that adequate immobilisation does not necessarily come at the cost of patient tolerance.

Setup accuracy has been rigorously quantified for head and neck targets. Sun et al. (2025) analysed 402 treatment fractions from 28 patients at the Shanghai Proton and Heavy Ion Center who received both supine and upright carbon ion radiotherapy. Median residual translational errors were less than 1 mm for both postures, and 78% of upright fractions fell within the 1 mm/1° threshold. The maximum calculated margin for upright was 1.8 mm, smaller than the 3.3 mm required for supine in the C1–C3 region. The authors attributed this to the comfort of the upright posture and the mandibular support provided by the mask.

Clinical Evidence: Where Upright Provides Advantage

The strongest clinical data come from head and neck and brain malignancies. Blumenfeld et al. (2025) reported the first prospective trial of gantry-less upright proton therapy for locally recurrent head and neck and brain cancers. Twenty patients were treated using the P-Cure system with daily vertical CT-based positioning. At three months, 64.3% of evaluable patients showed a complete response and 21.4% a partial response, with no disease progression. Dosimetric comparisons against supine photon IMRT showed significant reductions in brainstem Dmax (6 Gy[RBE] vs 19.1 Gy), spinal cord Dmax (12.5 vs 18.3 Gy), and oral cavity mean dose (5.6 vs 18.5 Gy). A four-modality substudy comparing upright PT, supine PT, upright IMRT, and supine IMRT found that upright PT matched or exceeded supine PT for OAR sparing in three of four representative patients, with reductions of 23% in mandible mean dose and 18% in oral cavity mean dose for one complex case.

The paediatric context is especially compelling. Blumenfeld et al. (2026) described the first paediatric case of upright proton craniospinal irradiation (CSI) under monitored anaesthesia care. A 4-year-old with relapsed CNS neuroblastoma completed 17 fractions without interruption, with no grade ≥2 acute toxicities. Mean translational shifts were 2.4 mm (cranial) and 2.1 mm (spinal), all within tolerance. Total session duration was 77 minutes initially, falling to 26 minutes for the boost phase, reflecting a clear workflow learning curve. While single-case evidence is limited, this proof-of-concept demonstrates that even complex paediatric geometries can be treated upright, potentially expanding access to proton CSI in settings without gantry-based facilities.

For thoracic targets, the evidence is more nuanced. Martire et al. (2026) compared upright and supine carbon ion plans for six lung cancer patients using 4D dosimetric analysis. Comparable target coverage (D95% ≈ 98%) was achieved for five of six patients when beam angles were optimised for the upright posture. However, direct translation of supine beam angles to upright increased OAR doses in five patients. The interplay effect favoured upright only in the patient with halved motion amplitude. The authors emphasised that the comparison is highly sensitive to beam angle selection, target contouring methodology, and assumed beamline flexibility. Importantly, when supine plans were constrained to fixed beamlines — as would be the case in a gantry-less carbon ion system — upright plans showed comparable or superior quality, with only a 3–9 percentage-point increase in lung V16Gy.

Technical innovations may further tip the balance. Engwall et al. (2025) demonstrated that combining upright static proton arcs with shoot-through layers — where protons traverse the patient and exit — can sharpen the beam penumbra without requiring collimation. In a head and neck case, shoot-through arcs achieved a conformity index of 0.48 versus 0.41 for 5-beam IMPT, with an estimated delivery time of 5.4 minutes. The method also enables straightforward range verification using a fixed detector opposite the nozzle, addressing a key concern in upright proton delivery.

Limitations and Unsuitable Indications

Despite these advances, upright positioning is not a universal solution. The SWOT analysis by Underwood et al. (2025) identified several classes of patients and treatments where upright may be disadvantaged. Patients requiring general anaesthesia with supine positioning, those with severe kyphosis or scoliosis, and very obese patients represent logistical challenges for chair-based systems. Non-coplanar fields are difficult to implement because tilting the patient alters internal anatomy, requiring additional imaging and potentially diminishing the simplicity advantage of a fixed beamline.

For pelvic and abdominal targets, the evidence base is thinner. Immobilisation studies by Varnava et al. (2025) showed that upright abdominal setups still produce interfractional errors of 5.5 mm even with shell immobilisation — roughly double the head and neck errors. While small photon series have shown patient preference for upright pelvic treatments, comprehensive planning studies comparing tumour control probabilities and normal tissue complication probabilities for upright versus recumbent positions are lacking for prostate, gynaecological, and lower gastrointestinal indications.

The Martire et al. (2026) lung study also revealed a potential limitation: target contouring bias. For one patient, using originally contoured versus propagated ITVs changed heart V20Gy by 11 percentage points — a difference that outweighed the true postural effect. This highlights the absence of dedicated upright imaging protocols for target delineation. Until upright MRI or PET becomes widely available, reliance on supine-to-upright deformable registration introduces uncertainties that may be unacceptable for stereotactic or ablative treatments.

Furthermore, the Blumenfeld et al. (2025) trial reported that 70% of patients required adaptive replanning, triggered by tumour progression or anatomic change. While adaptive workflows are feasible with daily upright CT, they add complexity that may offset some of the throughput advantages. The authors also noted that the conformity index was modestly lower for upright PT (0.76–0.87 versus 0.98 for supine IMRT) in complex geometries, suggesting that gantry-based systems may still be necessary for targets with highly irregular shapes or close proximity to multiple serial OARs.

Configuration Implications: Gantry vs Upright

Underwood et al. (2025) estimated that gantry-less upright systems reduce treatment room costs by a factor of 3–5 for protons — from approximately $30–50 million for a single-room gantry system to $10 million for a fixed-beamline upright room. The concrete shielding requirement drops from ~200 m³ to ~60 m³, with a corresponding 60-tonne reduction in CO₂ emissions per bunker (Volz et al., 2024). For carbon ion therapy, the savings are even more dramatic, given that carbon gantries require bunkers up to 25 m in length and three storeys in height.

These economics invite a rethinking of facility design. A single-room proton therapy system treating a diverse caseload — from simple prostate to complex multi-field paediatric head and neck — will almost certainly require a gantry. No evidence yet suggests that upright delivery alone can match the beam-angle flexibility and plan quality of a modern superconducting gantry for the most demanding cases. However, the evidence from Blumenfeld et al. (2025), Sun et al. (2025), and Martire et al. (2026) demonstrates that upright treatment can handle a substantial proportion of indications — particularly cranial, head and neck, and potentially thoracic — with comparable accuracy and dosimetry. This opens the door to a hybrid configuration: one gantry room serving the minority of patients who truly need multi-angle beam delivery, and two or three upright rooms handling the majority of straightforward head and neck, re-irradiation, and elective treatments.

Such a configuration would preserve the gantry’s role for complex geometries while capturing the cost and footprint advantages of upright rooms for the bulk of the patient population. The paediatric CSI case reported by Blumenfeld et al. (2026) shows that even a demanding technique can be delivered upright, potentially reducing the number of gantry slots needed in a paediatric centre. The shoot-through arc method from Engwall et al. (2025) further suggests that upright delivery quality can be improved without hardware additions, making upright rooms more versatile over time.

The main barriers to this hybrid model are not technical but organisational: staff training, workflow integration, and the need to maintain two parallel immobilisation and imaging pipelines. The SWOT analysis (Underwood et al., 2025) rightly notes that historically, in-house upright systems have been trialed and abandoned, often due to lack of upright CT scanners or resistance to deviating from optimised supine workflows. The availability of commercial integrated chair-CT systems and the accumulating clinical evidence may now be sufficient to overcome these barriers.

Conclusion

The evidence from the past three years establishes that upright particle therapy is not merely a cost-saving curiosity but a clinically viable alternative for selected indications. For cranial, head and neck, and re-irradiation cases, the dosimetric and positional data meet or exceed the supine standard. For lung cancer, the benefits are more patient-specific and require careful beam angle optimisation, but are real. For pelvic and abdominal targets, the data remain too sparse to recommend upright positioning outside clinical trials.

The optimal facility configuration is likely a hybrid: one gantry room for the most complex cases and multiple upright rooms for the majority of patients. This layout balances cost, footprint, and clinical breadth, and is supported by the emerging evidence that upright delivery can handle all but the most geometrically challenging treatments. The next steps are clear: larger prospective trials that report long-term local control and late toxicity, the development of upright immobilisation solutions for the pelvis and abdomen, and the integration of upright imaging modalities — particularly MRI — to improve target delineation. If these are achieved, the gantry may cease to be the universal workhorse of particle therapy and instead become a specialized tool reserved for the cases that truly require it.

References

Volz et al. Considerations for Upright Particle Therapy Patient Positioning and Associated Image Guidance. · PMID: 35965576

Volz et al. Opportunities and challenges of upright patient positioning in radiotherapy. · PMID: 39159668

Sun et al. Comparative analysis of residual setup errors in head and neck patients from upright versus supine radiotherapy postures. · PMID: 40216595

Underwood et al. A Strengths, Weaknesses, Opportunities, and Threats (SWOT) Analysis for Gantry-Less Upright Radiation Therapy. · PMID: 40278803

Blumenfeld et al. Prospective clinical trial of upright image-guided proton therapy for locally recurrent head and neck and brain malignancies. · PMID: 40803484

Blumenfeld et al. Pediatric CNS relapse of neuroblastoma treated with upright proton craniospinal irradiation. · PMID: 41362481

Varnava et al. Assessment of upright immobilization methods for abdominal and head-and-neck cancer treatments in a carbon ion radiotherapy setting. · PMID: 41376676

Engwall et al. Shoot-through layers in upright proton arcs unlock advantages in plan quality and range verification. · PMID: 40830045

Martire et al. Upright and supine particle therapy of lung cancer: A 4D dosimetric comparison. · PMID: 41776380