Continuous positive airway pressure in breast cancer radiotherapy: Protocol of a prospective clinical trial Authors: A Van der Vorst, J Verhoeven, A Baten, H Janssen, L Van Bavel, L Delombaerde, T Reynders, M Cooreman, A Laenen, M Lambrecht, Aelst L Van, C Weltens Published: 2026 ABSTRACT Graphical abstractUnlabelled ImageKeywords: Breast cancer, Radiation therapy, Continuous positive airway pressure, Toxicity, Dosimetry, Feasibility Introduction Breast cancer (BC) remains the most common cancer among women worldwide [1], [2]. Despite its high incidence, BC has a good prognosis with a five year overall survival (OS) of approximately 80–90 % [3]. Consequently, reducing long-term treatment related toxicity is essential. Treatment of early BC typically consists of surgery, radiotherapy (RT) and, when indicated, systemic therapy. Approximately 70 % of all BC patients undergo adjuvant RT as a part of their curative treatment [4]. However, BC RT is also associated with potential long-term toxicity, including pneumonitis, lung function reduction, second primary contralateral breast (CLB) and lung cancers, heart failure (HF), ischemia and rhythm disorders [5], [6], [7], [8], [9], [10], [11], [12], [13]. These side effects are uncommon, but can be severe or life-threatening. Most of these effects correlate with higher radiation doses to these organs-at-risk (OAR). Therefore, substantial effort is being put into the development of new techniques to reduce RT doses to the OARs. These include advances in RT techniques, such as intensity modulated RT (IMRT), volumetric modulated arc therapy (VMAT) and proton therapy (PT). In addition, breathing technologies like deep inspiration breath hold (DIBH) have become part of the standard of care (SOC) in the treatment of left-sided BC [9], [14]. A potential alternative option to reduce cardiac and pulmonary radiation doses is the use of CPAP (continuous positive airway pressure) during BC RT. A CPAP device delivers positive airway pressure in a continuous way (during inspiration and expiration), thereby opening the distal alveoli, leading to increased end-expiratory functional lung volumes. In BC RT, this could lead to a greater distance between the irradiated target volumes and the OARs, potentially reducing lung and heart doses since the total lung volume increases. Furthermore, by preventing alveolar collapse during exhalation, it may reduce work of breathing, therefore theoretically making DIBH maintenance easier in left-sided BC patients [15], [16]. Several studies in left-sided BC patients have evaluated CPAP added to free breathing (FB) or as an alternative to DIBH [17], [18], [19], [20], [21], [22]. These studies consistently demonstrated that CPAP increases lung volumes and reduces heart and lung doses while remaining well tolerated. In contrast, Choi et al., reported that CPAP at 10cmH2O achieved comparable dosimetric results compared to DIBH in terms of MHD, MLD, mean CLB dose and maximum LAD dose when VMAT was used as a planning technique [23]. Finally, one study compared CPAP-assisted DIBH with DIBH alone and confirmed superior OAR sparing with the combined approach, indicating that CPAP may function not only as an alternative to DIBH but also as an add-on that further enhances dosimetric outcomes compared to DIBH alone [24]. Table 1 presents an overview of CPAP studies in BC RT. Table 1Overview of CPAP studies in breast cancer radiotherapy.StudyDesignInclusionNumber of patientsRT techniqueOutcomeKilROJ2018Case reportLeft-sided PMRT + RNI Unable to perform DIBH13DCRTVMATMHD and mean ipsilateral lung dose lower with CPAPFB vs CPAP-FB12cmH2O KilAm J Clin Oncol2019Prospective, non-randomizedLeft-sided123DCRTMHD, mean ipsilateral lung dose, V20 lungs lower with CPAPFB vs CPAP-FB8-15cmH2O AllenRadiat Oncol2020Prospective, non-randomizedLeft-sided ± RNI173DCRTMHD, V20 lungs lower with CPAPFB vs CPAP-FB20.4cmH2O ReckhowMed Dosim2021RetrospectiveLeft-sided103DCRTMHD, MLD lower with CPAPFB vsCPAP-FBDIBHCPAP-DIBH15 cmH2O KoFront Oncol2021RetrospectiveLeft-sided + RNI15VMATPTMHD and MLD lowest in PT and VMAT-CPAPFB vs CPAP-FBCPAP pressure not specified ChoiPRO2022RetrospectiveLeft-sided20VMATMHD and mean ipsilateral lung dose lower with CPAPFB vs CPAP-FBCPAP12-15cmH2O KilAdv RO2024Prospective non-randomizedLeft-sided ± RNIUnable to perform DIBH103DCRTMHD and MLD lowest in CPAP-FBFB vsDIBHCPAP-FB15cmH2O ChoiAdvRTO2024RetrospectiveLeft-sided35VMATCPAP is comparable to DIBH for MHD, MLD, mean CLB and maximum LAD dose (both significantly better than FB)FB vsDIBHCPAP10cmH2OAbbreviations: 3DCRT: 3D Conformal Radiotherapy, CLB: Contralateral Breast, CPAP: Continuous Positive Airway Pressure, DIBH: Deep inspiration Breath Hold, FB: Free Breathing, LAD: Left Anterior Descendens, MHD: Mean Heart Dose, MLD: Mean Lung Dose, PMRT: Post-Mastectomy Radiotherapy, PT: Proton Therapy, RNI: Regional Nodal Irradiation, RT: radiotherapy, V20: Volume receiving 20 Gy, VMAT: Volumetric Modulated Arc Therapy. Overview of CPAP studies in breast cancer radiotherapy. Abbreviations: 3DCRT: 3D Conformal Radiotherapy, CLB: Contralateral Breast, CPAP: Continuous Positive Airway Pressure, DIBH: Deep inspiration Breath Hold, FB: Free Breathing, LAD: Left Anterior Descendens, MHD: Mean Heart Dose, MLD: Mean Lung Dose, PMRT: Post-Mastectomy Radiotherapy, PT: Proton Therapy, RNI: Regional Nodal Irradiation, RT: radiotherapy, V20: Volume receiving 20 Gy, VMAT: Volumetric Modulated Arc Therapy. Most of these studies however were rather small, only investigated its use in left-sided BC patients and only three studies compared its use against current SOC (DIBH). Therefore, the objective of this larger, prospective clinical trial is to investigate whether the addition of CPAP to SOC improves OAR dosimetry in both left- and right-sided patients and to identify patient subgroups that benefit most from CPAP-assisted RT. Furthermore, feasibility will be assessed through patient questionnaires, treatment time, accuracy and reproducibility in clinical practice. In addition, DIBH duration and depth will be evaluated. Ultimately, this study aims to establish patient selection criteria and provide guidelines for clinical implementation of CPAP in BC RT. Methods/Design Study design This prospective clinical trial investigates the potential dosimetric benefits and clinical feasibility of incorporating CPAP in BC RT. It is a single arm study in which patients undergo SOC computed tomography (CT) simulation followed by an additional CT simulation with the addition of CPAP to SOC. For left-sided patients, SOC consists of DIBH. For right-sided patients and left-sided patients unable to perform DIBH or aged ≥70 years, SOC consists of FB (the rationale for the DIBH age threshold of <70 years is provided in Supplementary Material 1). For each patient, two RT treatment plans are generated: with and without the addition of CPAP. RT plans are compared for target coverage and OAR doses. If CPAP reduces mean heart or CLB dose by ≥0.5 Gy, or mean lung dose (MLD) by ≥1 Gy, treatment is delivered with CPAP. A single-arm design was selected because interpatient anatomic variation strongly influences dose distributions. A randomized comparison between CPAP and no CPAP across different patients would risk confounding anatomical differences with treatment effects. By comparing CPAP-assisted and SOC plans within the same patient, we ensure that any observed differences can be attributed to CPAP rather than to anatomical variation. An overview of the study design is presented in Fig. 1. Study design. Abbreviations: BCS: Breast Conserving Surgery, DIBH: Deep Inspiration Breath Hold, CLB: Contralateral Breast, CPAP: Continuous Positive Airway Pressure, CT: Computed Tomography, Gy: Gray, MHD: Mean Heart Dose, MLD: Mean Lung Dose, RNI: Regional Nodal Irradiation, RT: Radiotherapy, SOC: Standard Of Care, y/o: Year Old. Inclusion criteria Women aged 40–80 years old requiring breast conserving surgery (BCS) and RT for invasive or in situ BC are prospectively enrolled. Eligible cases include left-sided disease with or without regional nodal irradiation (RNI) and right-sided disease requiring RNI. Exclusion criteria Contraindications include comorbidities precluding CPAP use or RT, pregnancy, inadequate arm mobility, or prior overlapping RT. Additional exclusion criteria include: post-mastectomy RT, nodal boost, distant metastases, breast implants, bilateral BC, concomitant chemotherapy, other active malignancy (excluding non-melanoma skin cancer), and the requirement for an electron boost. Treatment description CPAP equipment Fisher & Paykel Evora Full Face masks are used, with a Philips DreamStation1 CPAP device. CT simulation The CPAP mask is first positioned with adequate coverage of the nose and mouth, without activating the CPAP device. The patient is then placed in RT position: supine on a breast board with arms elevated. Standard non-contrast CT imaging is performed first—consisting of a voluntary DIBH scan for left-sided patients, and a FB scan for right-sided patients or left-sided patients unable to perform DIBH or ≥70 years old. Next, the CPAP device is activated, starting at a pressure of 7 cm H2O increasing to 15 cm H2O over 10 min. Once the target pressure is reached, an additional 5-min stabilization period follows to allow full lung expansion. Thereafter, a CPAP-assisted CT scan is acquired: in DIBH for left-sided patients, and in FB for right-sided patients or left-sided patients unable to perform DIBH or ≥70 years old. Delineation The breast and nodal clinical target volumes (CTV) are delineated according to ESTRO guidelines [25]. The boost volume is contoured following Boersma et al. [26]. The OARs delineated include the heart, cardiac substructures, lungs, CLB, the esophagus in cases involving RNI, and the liver for right-sided RT. Cardiac substructures comprise the ascending aorta, superior vena cava, pulmonary artery, cardiac chambers, conduction nodes, cardiac valves, coronary arteries, and the heart base [27], [28], [29]. Radiotherapy planning RT planning is performed for both the SOC and CPAP-assisted CT scans. First planning target volumes (PTV) are generated by expanding the delineated CTVs by 7 mm for the breast and 5 mm for the boost volume and/or RNI. Photon-based RT is planned using 6FFF energy, prescribing 40.05 Gy to the breast and nodes, with a simultaneous integrated boost (SIB) of 48 Gy to the boost volume in 15 fractions of 2.67 Gy and 3.2 Gy, respectively. Breast and RNI volumes are planned using IMRT, whereas the boost is delivered using VMAT. Both SOC and CPAP plans are generated using comparable field arrangements and optimized to achieve similar target coverage: at least 95 % of the breast and boost PTVs must receive ≥95 % of the prescribed dose, and 95 % of the RNI PTV must receive ≥85 % of the prescribed dose. The maximum dose to any volume must remain ≤ 107 % of the prescription. Dose constraints and clinical target coverage goals are detailed in Supplementary Material 2. Selection of best radiotherapy treatment plan Target coverage and dose constraints must be met for both RT plans, although minor deviations are permitted if approved by the treating physician. If CPAP has similar or better target coverage and reduces MHD or CLB dose by ≥0.5 Gy, or MLD by ≥1 Gy, treatment is delivered with CPAP. Selection of tresholds Reduction of MHD by at least 0.5 Gy Several small, earlier studies have demonstrated that the addition of CPAP to BC RT is associated with reductions in MHD [17], [18], [19], [20], [21], [22], [23], [24]. Based on extrapolation of the reported MHD reductions in these studies, combined with the baseline MHD observed in our BC population at our center (mean 1.49 ± 0.70 Gy), we anticipated that a relative MHD reduction of approximately 30–50 % would be achievable. Such a reduction would not only be technically feasible but also clinically meaningful, as multiple studies have established a linear relationship between MHD and the risk of cardiac toxicity—particularly coronary artery disease—with relative risk increases of up to 19 % per Gy increase in MHD [30]. Accordingly, a reduction of at least 0.5 Gy in MHD was considered clinically relevant. Reduction of mean CLB dose by at least 0.5 Gy As stated by the Dutch selection model for PT, the absolute excess risk of developing a second primary contralateral BC increases by approximately 2.5 % per Gy increase in mean CLB dose [31]. Furthermore, considering established dose constraints −typically recommending a mean CLB dose below 4 Gy, and even below 2 Gy at our institution- lead to the decision that a reduction of 0.5 Gy in mean CLB dose is both clinically meaningful and potentially achievable with the addition of CPAP [32]. Reduction of MLD by at least 1 Gy Boersma et al. reports an absolute excess risk of approximately 11 % per Gy increase in MLD for the development of second primary lung cancer, depending on gender and smoking status [31]. In addition, higher lung doses are associated with an increased risk of radiation pneumonitis. Karlsen et al. demonstrated that a higher MLD was significantly associated with the incidence of radiation pneumonitis at six months following completion of RT [33]. Moreover, given that clinical constraints for MLD typically aim to remain below 6 Gy, a reduction of at least 1 Gy in MLD was therefore considered clinically meaningful [32]. Radiotherapy delivery Patients are positioned on the treatment coach in FB using surface scanning. Daily cone beam CT (CBCT) is performed – either in FB or in DIBH, depending on tumor laterality and patient age. Image registration is initially performed automatically and subsequently refined manually when necessary. Matching is based on the breast surface, surgical clips, and seroma boundaries when present. When CPAP-assisted RT is used, the CPAP mask is first fitted to the patient’s face prior to positioning, without activating the device. Once final positioning is confirmed, the CPAP device is activated and airway pressure is increased from 7 cm H2O to 15 cm H2O over 10 min, followed by an additional 5-min stabilization period. CBCT imaging is then acquired under CPAP assistance—either in FB or DIBH, depending on tumor laterality and patient age—using the same image registration workflow and matching structures as applied in the SOC setting. Endpoints All study endpoints will be assessed for the overall cohort and for 5 subgroups of BC patients: right-sided with RNI, left-sided without RNI, left-sided with RNI (<70y/o), left-sided with RNI (>70 years old), left-sided unable to perform DIBH. Primary endpoint The primary endpoint of this study is absolute mean heart dose (MHD) reduction with CPAP-assistance, expressed in Gy. Secondary endpoints Secondary endpoints are: (1)Comparison of clinical factors between patients that underwent standard versus CPAP-assisted RT-Age-Body Mass Index-Cardiovascular (CV) risk factors (smoking, diabetes, arterial hypertension, CV history)-Oncological data-BC laterality (left versus right) and sublocalisation in the breast quadrant-Clinical and pathological TNM stage-Tumor size, subtype, grade, estrogen receptor (ER) progesterone receptor (PR) status, HER2-status, Ki67, management of the axilla, neoadjuvant and adjuvant systemic therapy-Extent of the RT field (with or without RNI)-RT with or without DIBH Comparison of clinical factors between patients that underwent standard versus CPAP-assisted RT-Age-Body Mass Index-Cardiovascular (CV) risk factors (smoking, diabetes, arterial hypertension, CV history)-Oncological data-BC laterality (left versus right) and sublocalisation in the breast quadrant-Clinical and pathological TNM stage-Tumor size, subtype, grade, estrogen receptor (ER) progesterone receptor (PR) status, HER2-status, Ki67, management of the axilla, neoadjuvant and adjuvant systemic therapy-Extent of the RT field (with or without RNI)-RT with or without DIBH -Age Age -Body Mass Index Body Mass Index -Cardiovascular (CV) risk factors (smoking, diabetes, arterial hypertension, CV history) Cardiovascular (CV) risk factors (smoking, diabetes, arterial hypertension, CV history) -Oncological data Oncological data -BC laterality (left versus right) and sublocalisation in the breast quadrant BC laterality (left versus right) and sublocalisation in the breast quadrant -Clinical and pathological TNM stage Clinical and pathological TNM stage -Tumor size, subtype, grade, estrogen receptor (ER) progesterone receptor (PR) status, HER2-status, Ki67, management of the axilla, neoadjuvant and adjuvant systemic therapy Tumor size, subtype, grade, estrogen receptor (ER) progesterone receptor (PR) status, HER2-status, Ki67, management of the axilla, neoadjuvant and adjuvant systemic therapy -Extent of the RT field (with or without RNI) Extent of the RT field (with or without RNI) -RT with or without DIBH RT with or without DIBH (2)Comparison of RT doses to the OARs between standard and CPAP protocol-Absolute mean dose (Dmean) (Gy)-Absolute maximum dose (Dmax) (Gy)-Volumetric parameters (V5Gy, V10Gy, V15Gy, V20Gy, V30Gy, V40Gy, V50Gy, V60Gy) Comparison of RT doses to the OARs between standard and CPAP protocol-Absolute mean dose (Dmean) (Gy)-Absolute maximum dose (Dmax) (Gy)-Volumetric parameters (V5Gy, V10Gy, V15Gy, V20Gy, V30Gy, V40Gy, V50Gy, V60Gy) -Absolute mean dose (Dmean) (Gy) Absolute mean dose (Dmean) (Gy) -Absolute maximum dose (Dmax) (Gy) Absolute maximum dose (Dmax) (Gy) -Volumetric parameters (V5Gy, V10Gy, V15Gy, V20Gy, V30Gy, V40Gy, V50Gy, V60Gy) Volumetric parameters (V5Gy, V10Gy, V15Gy, V20Gy, V30Gy, V40Gy, V50Gy, V60Gy) (3)Determination of patient comfort and tolerance of CPAP protocol by analysing patient questionnaires (Likert scores from patient questionnaires, see Supplementary Material 3.) Determination of patient comfort and tolerance of CPAP protocol by analysing patient questionnaires (Likert scores from patient questionnaires, see Supplementary Material 3.) (4)Comparison of time-effectiveness between standard and CPAP-assisted RT: set-up, matching and radiation time (minutes). Comparison of time-effectiveness between standard and CPAP-assisted RT: set-up, matching and radiation time (minutes). (5)Comparison of reproducibility, stability and accuracy between standard and CPAP-assisted RT under both FB and DIBH conditions by using surface scanning parameters and daily CBCT information acquired during treatment. Comparison of reproducibility, stability and accuracy between standard and CPAP-assisted RT under both FB and DIBH conditions by using surface scanning parameters and daily CBCT information acquired during treatment. (6)Comparison of DIBH performance in left-sided patients between standard and CPAP protocol: DIBH duration (seconds) and depth (centimeters). Comparison of DIBH performance in left-sided patients between standard and CPAP protocol: DIBH duration (seconds) and depth (centimeters). Statistics Hypotheses The primary hypothesis of this study is that CPAP-assisted RT leads to a reduction in MHD. Secondary hypotheses are that CPAP-assisted RT results in dose reductions to other OARs (lungs, CLB, cardiac substructures, and the liver), improves DIBH performance, is associated with acceptable patient comfort, does not significantly increase treatment device time, and provides reproducible and accurate treatment delivery. Sample size The aim is to evaluate whether CPAP-assisted RT improves cardiac radiation doses in 3 main cohorts of BC patients (right-sided BC, left-sided BC without RNI, left-sided BC with RNI). A decrease of 0.5 Gy MHD by using CPAP-assisted RT compared to a standard protocol was deemed clinically relevant. Retrospective data provided the following information on means and standard deviation (SD) of MHD in the 3 cohorts: right-sided BC Mean MHD 1.03 Gy, SD 0.52 Gy; left-sided BC without RNI Mean MHD 1.15 Gy, SD 0.71 Gy; left-sided BC with RNI Mean MHD 2.29 Gy, SD 0.82 Gy. No information was available on the correlation between paired measurement of MHD with and without CPAP. A correlation of 0.5 was assumed for the sample size calculation, leading to sample sizes of 11 (right-sided BC), 18 (left-sided BC without RNI), and 24 (left-sided BC with RNI) patients, or 53 patients in total. The sample size calculation was performed assuming a two-sided paired t-test, adopting a 5 % test-wise significance level, and 80 % of power. For the additional subgroup of patients aged 70–80 years with left-sided BC and an indication for RNI, recruitment is expected to be limited. Similarly, the subgroup of left-sided patients unable to perform DIBH is anticipated to be small, since most younger and fit patients can perform DIBH. Consequently, no formal sample size calculation was performed for these two subgroups, and their results will be considered exploratory. In contrast, for the three primary subgroups, a total sample size of approximately 53 patients is anticipated. Data analysis For each of the five cohorts, dosimetric and time-efficiency outcomes will be analyzed using two-sided paired t-tests to compare differences between the standard protocol and CPAP. In left-sided patients, DIBH performance will additionally be compared between the standard and CPAP protocol using two-sided paired t-tests. The mean difference will be estimated with 95 % confidence interval. A 5 % significance level will be adopted. Patient comfort and tolerance of CPAP-assisted RT assessed by patient questionnaires and will be analyzed descriptively, including frequencies, percentages, means, and standard deviations (or medians and interquartile ranges, as appropriate). A Bland-Altman analysis will be performed to analyze agreement of surface scanning parameters obtained with CPAP or standard protocol. The focus of the secondary analyses will be on the estimation of effect sizes with 95 % confidence intervals. Formal tests performed will adopt a 5 % significance level, nevertheless p-values will be interpreted carefully, recognizing the risk inflated type I error rate. Conclusions will be drawn as hypothesis-generating. We expect negligible missing observations as all outcome measurements are collected prospectively by clinical staff. Planned timeline Patient inclusion began in February 2025 and is expected to be completed by December 2026. Declaration of generative AI and AI-assisted technologies in the manuscript preparation process During the preparation of this work the author(s) used ChatGPT in order to improve linguistics solely. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article. Author contribution Conceptualization and design: AVDV, CW, ML, LVA. Study coordination: AVDV, CW. Patient recruitment and data acquisition: AVDV, JV, AB, HJ, LVB, MC. Statistical analysis: AL. Interpretation of data: AVDV, CW, ML, LVA, AL. RT planning: LD, TR. Data extraction from RT software: AVDV, LD. Supervision: CW, ML, LVA. Drafting the manuscript: AVDV, CW. Revision and approval of the manuscript: all authors. Accountable of all aspects of the work: all authors. All authors read and approved the final manuscript. Ethics approval and consent to participate This study was approved by the Ethical committee form UZ/KU Leuven: S68702 (Belgian registration number B3222024001447). The study is also registered on Clinicaltrials.gov: NCT07373782. Written informed consent was obtained from all patients prior to study enrollment. Funding During the preparation of this manuscript, Dr. Aline Van der Vorst has been funded by Kom op tegen Kanker (Stand up to Cancer), the Flemish cancer society (projectID:13905). The CPAP masks used in this study were kindly donated by VitalAire/Air Liquide Homecare. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements We thank Eszter Hortobágyi, Laure Pape, Quinten Goessaert and Alec Bomhals for their contributions to the RT planning in this study. Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.ctro.2026.101209. Supplementary data The following are the Supplementary data to this article: Supplementary Data 1 Supplementary Material. (1) Rationale for the age threshold of <70 years old for performing deep inspiration breath hold (DIBH). Supplementary Data 2 (2) Clinical goals and dose constraints. Supplementary Data 3 (3) Patient comfort questionnaire. Supplementary Data 4 (4) Study protocol. References Associated Data Supplementary Materials Supplementary Data 1 Supplementary Material. (1) Rationale for the age threshold of <70 years old for performing deep inspiration breath hold (DIBH). Supplementary Data 2 (2) Clinical goals and dose constraints. Supplementary Data 3 (3) Patient comfort questionnaire. Supplementary Data 4 (4) Study protocol.