Advances in childhood cancer treatment and supportive care have resulted in survival rates that now exceed 80% in high-income countries [1]. Yet, even as cancer treatments are curative in targeting cancer cells, they can harm healthy tissue and potentially cause late complications, such as second neoplasms and chronic diseases [2,3]. The cumulative incidence of such late effects among childhood cancer survivors (CCS) thus predisposes them to increased morbidity and premature mortality [4]. Among late effects, pulmonary complications are the third-leading cause of excess mortality after second neoplasms and cardiovascular diseases [5].

Several cancer treatment modalities can be pulmotoxic. These include the chemotherapeutic agents busulfan, bleomycin, carmustine, and lomustine; thoracic radiotherapy; thoracic surgery; and hematopoietic stem cell transplantation (HSCT) [6,7]. The underlying mechanisms of pulmonary toxicity involve alveolar, vascular, and parenchymal damage resulting from chemotherapy and radiotherapy, which may progress to lung fibrosis [6-8]. HSCT-related lung damage can result from intensive conditioning regimens, infections, or graft-versus-host disease (GvHD), while surgery of the chest or lungs may reduce lung volumes and impair chest wall compliance [7,9]. However, recently published recommendations for surveillance of pulmonary function among CCS from the International Late Effects of Childhood Cancer Guideline Harmonization Group (IGHG) could not consistently confirm the pulmonary toxicity of all these treatments, particularly certain chemotherapeutics, due to limited and low-quality evidence [10]. Other chemotherapeutics, such as methotrexate and cyclophosphamide, have also been suspected of causing lung damage [8,11], but findings across studies remain inconsistent [12,13]. Considering that these treatments may harm developing lungs and potentially lead to progressive pulmonary damage [14], it is important to study their effects on lung function.

Pulmonary dysfunction in CCS exposed to pulmotoxic treatments has been reported in varying proportions of CCS (44%-77%), depending on study populations and criteria used to define obstructive, restrictive, and diffusion impairments [12,13,15,16]. The lung has a large functional reserve, and early disease may often remain asymptomatic, particularly when it affects the lung periphery [17]. Most studies have used conventional pulmonary function tests (PFTs), such as spirometry, body plethysmography, and diffusing capacity of the lung for carbon monoxide (DLCO), to assess lung function in CCS. However, spirometry and body plethysmography lack the sensitivity to detect changes in small airways [18], which may be damaged first [6]. The nitrogen multiple breath washout test (N₂MBW), which measures ventilation inhomogeneity of the ventilated lung, detecting small airway disease, has been increasingly used in other patient populations, including those treated with allogeneic HSCT [19-21]. In a small prospective study of adult CCS, N₂MBW identified more cases of pulmonary dysfunction than spirometry, even among those who had not been exposed to previously defined pulmotoxic treatments [22]. Though larger prospective studies with standardized assessments are still needed, these findings suggest that N₂MBW could be a valuable complementary test in screening CCS for early pulmonary damage.

The recently published IGHG recommendations not only summarized existing evidence but also highlighted large knowledge gaps and methodological weaknesses in previous research [10]. These gaps emphasize the need for prospective, longitudinal studies with larger sample sizes and a broader range of treatment exposures to characterize the onset and progression of treatment-related pulmonary dysfunction. The long-term effects of newer chemotherapeutic and immunotherapeutic agents on lung function are understudied [23]. Evidence on how treatment-related complications, comorbidities, and genetic variants influence lung damage risk is limited as well. Similarly, standardization is lacking in PFTs and in the use of appropriate age- and sex-specific reference values. For example, results should be reported as z-scores rather than just proportions of patients with reduced lung function. Additionally, few studies have assessed diagnostic tests specific to the location and type of potential dysfunction, such as N₂MBW for peripheral inhomogeneity in ventilation. More data are also needed on the association between PFT outcomes and clinical symptoms. To address these gaps, we designed the Swiss Childhood Cancer Survivor Study (SCCSS) FollowUp–Pulmo.

The primary objective of SCCSS FollowUp–Pulmo is to longitudinally investigate lung function in children and adolescents after cancer treatment using a comprehensive set of PFTs that also assess small-airway disease. Second, the study will investigate possible effects of treatment-related risk factors (systemic anticancer agents, thoracic radiotherapy, thoracic surgery, and HSCT), treatment-related complications (pulmonary infections, GvHD), and existing comorbidities (eg, pulmonary or cardiac disease) on lung function. Third, the study will examine the ability of N₂MBW to detect pulmonary dysfunction in comparison with other PFTs. Fourth, it will investigate the association between lung function and self-reported respiratory symptoms.

SCCSS FollowUp–Pulmo is a multicenter prospective longitudinal cohort study integrated into the routine clinical care of several children’s hospitals in Switzerland. The study is an interdisciplinary collaboration between the pediatric hematology/oncology and pediatric pulmonology departments of respective centers, the Institute of Social and Preventive Medicine (ISPM) at the University of Bern, and the Swiss Childhood Cancer Registry (ChCR). The ChCR is a nationwide, population-based cancer registry that includes Swiss residents diagnosed up to the age of 20 years with leukemia, lymphoma, central nervous system (CNS) tumors, malignant solid tumors, or Langerhans cell histiocytosis, classified according to the International Classification of Childhood Cancer, Third Edition (ICCC-3) [24]. Although the ChCR captures over 95% of children diagnosed in Switzerland [25], SCCSS FollowUp–Pulmo will also include a few patients treated and followed up in Swiss clinics who may not be registered in the ChCR due either to registration delays or residency outside of Switzerland.

All CCS aged 6-20 years for whom 1 year or more has elapsed since cancer diagnosis, who have completed treatment, and who are in regular pediatric hemato-oncological follow-up care are eligible for SCCSS FollowUp–Pulmo. The following treatment modalities possibly affecting lung function are included in SCCSS FollowUp–Pulmo: any systemic anticancer treatment (chemotherapy, immunotherapy, or targeted therapy) [26]; thoracic surgery involving the chest or lungs (excluding central line placement) [27]; radiation of the lungs, the chest (axilla, mantle, mediastinal), or scattered radiation from other radiation fields, including the whole abdomen or any upper abdominal field, spinal doses of 30 Gray or higher, and total body irradiation [27]; and HSCT [28]. Excluded are CCS who were treated only with surgery or radiation outside the thorax due to their low risk for pulmonary dysfunction [29] and patients with relapse or in palliative care at the time of recruitment.

Ethical approval was granted by the Ethics Committee of the Canton of Bern, Switzerland (KEK-BE: 2019-00739), and the study is registered on ClinicalTrials.gov (NCT04732273). All participants provide a signed informed consent form. If any patient declines to participate, their data are not collected. Data are entered into the Research Electronic Data Capture (REDCap) database version 14.0.10 (Vanderbilt University), which complies with legal requirements for data security and data protection.

The study is coordinated by the ISPM research team, which is responsible for overall study management, monitoring recruitment progress, and handling administrative and financial aspects. Clinical teams at each participating center include pediatric oncologists, pulmonologists, data managers, and study nurses.

To determine the sample size needed for our study, we based calculations on the study by Schindera et al [22], who assessed lung function in long-term CCS using spirometry and N₂MBW. This single-center study was conducted on Swiss CCS and applied the same inclusion criteria for treatment exposures, making it a suitable reference. The main outcome from N₂MBW was the lung clearance index (LCI). The mean LCI z-score was 1.37 (SD 2.69). We calculated the number of participants necessary to achieve a statistical significance level of 0.05 and a power of 0.80, while accounting for a 15% dropout rate. This calculation indicated that a minimum of 146 participants would be required to detect a similar deviation in LCI with sufficient statistical power. We plan to include a larger sample size to increase precision, improve our ability to detect smaller deviations, and ensure adequate power to analyze other lung function outcomes and investigate specific subgroups of patients defined by tumor type and treatments received.

To compare PFT outcomes with normal values, we will calculate z-scores and percent-predicted values using external Global Lung Initiative (GLI) reference data [35-37,39]. We will define pulmonary dysfunction as z-scores below –1.64 or above +1.64 for respective PFT indices since these thresholds represent deviations from the reference population mean and indicate abnormalities in pulmonary function. To characterize pulmonary function among CCS, we will analyze z-scores for predefined outcomes and assess differences based on characteristics such as treatment exposure, age at treatment, and time since cancer diagnosis. We will assess group differences using appropriate statistical tests based on the type of variable, such as the t-test or the Mann-Whitney test for continuous variables and the χ² test or the Fisher exact test for categorical variables. For analyses of associations between outcomes and covariates, we will apply regression models adjusted for potential confounders. For longitudinal data, we will use mixed effects models to account for repeated measures over time. We will use Stata (Stata Corp LLC) and R (Foundation for Statistical Computing) for statistical analyses.

This prospective, multicenter cohort study of lung function in children and adolescents after cancer treatment investigates risk factors, compares the detection of pulmonary dysfunction using N₂MBW and conventional PFTs, and examines the association of functional outcomes from PFTs with respiratory symptoms.

Previous studies of pulmonary dysfunction after treatment for childhood cancer have mostly been retrospective, based on chart reviews [14,40-42]. This has entailed methodological weaknesses, particularly a risk of selection bias because CCS for whom PFT results were obtained might overrepresent symptomatic CCS or those treated more intensively. We identified several prospective studies on pediatric CCS, but like the retrospective ones, they mostly recruited specific groups of patients treated with previously defined pulmotoxic treatments, such as HSCT [42-44] or thoracic radiation [14,45], or specific tumor types like leukemia [46] or lymphoma [47,48]. Data on the pulmotoxic effects of individual agents, such as busulfan [49], melphalan [50], cyclophosphamide [46,50], and methotrexate [51], remain limited. By including CCS exposed to systemic therapies, including any chemotherapy, targeted agents, or immunotherapy, our study will provide a better understanding of treatment-related pulmonary dysfunction in a broadly representative population of CCS. Continuous recruitment will allow us to collect data on the pulmonary effects of newer treatments used in contemporary protocols, whose impacts remain largely unknown [52]. Detailed information from medical records will allow the investigation of effects of comorbidities and treatment-related complications on PFT outcomes.

Previous studies have rarely included sensitive tests, such as N₂MBW, that can detect early changes in the lung periphery. Most have used spirometry, body plethysmography, and DLCO [12,14,15,49]. A study investigating 57 pediatric CCS with a median follow-up time of 6.2 years from end of treatment did not find differences in ventilation inhomogeneity measured using N₂MBW compared to healthy controls [53]. In contrast, several studies on patients after allogeneic HSCT have reported the LCI to be a sensitive measure for early pulmonary complications [20,21,54]. This study will obtain N₂MBW data for CCS exposed to a wide range of treatment modalities and investigate whether N₂MBW is more sensitive than other PFTs in detecting early pulmonary dysfunction.

Another drawback of many existing studies is that they report PFT data using binary cut-offs and describe results as either normal or abnormal [12,15,49]. This reduces statistical power and introduces interpretations based on predefined threshold values. Reporting PFT results as raw data and z-scores based on internationally agreed-upon, age-adjusted reference values will allow better comparison and pooling across studies. Limited data exist on how lung function correlates with clinical symptoms [10]. Questionnaire data collected in this study will help investigate symptoms and other patient-reported outcomes, as well as their correlation with PFT results.

The SWISS-Pearl Study (ClinicalTrials.gov ID: NCT05427136), currently conducted in multiple centers in Switzerland, investigates lung function in patients with pediatric cancer. The study includes spirometry, body plethysmography, N₂MBW, magnetic resonance imaging of the lungs, and questionnaires at different points during cancer treatment. We plan to combine that study with ours to create a comprehensive database, enabling us to analyze lung function trajectories in patients and survivors of pediatric cancer.

The GECCOS (Genetic Risks for Childhood Cancer Complications Switzerland) study is a nationwide cohort study collecting germline genetic data from patients and survivors of childhood cancer in Switzerland [55]. In consenting patients, we will link clinical and PFT data with the genetic data from GECCOS, allowing us to investigate the effects of genetic predisposition on pulmonary toxicity. This will assist the development of personalized treatment strategies and risk-adapted long-term care for survivors.

A current limitation of this study is the limited number of participants and the heterogeneity of the study population, which can limit the statistical power for conducting subgroup analyses. For instance, exploring rare tumors or assessing specific effects of individual chemotherapeutic agents may at present be challenging. However, we plan to expand the study to more centers and pool data with international collaborations. Another limitation is the lack of systematic baseline PFT assessments before cancer treatment, making it difficult to distinguish treatment-related pulmonary dysfunction from preexisting conditions. Future studies should include pretreatment PFTs to better track changes in z-scores over time and improve the assessment of therapy-related effects. Finally, there is a small risk of selection bias because CCS with a longer time since the end of treatment, who do not experience respiratory symptoms or were not exposed to pulmotoxic treatments, may be less likely to participate in the study, potentially leading to underrepresentation of healthy CCS. Yet, because the study is embedded in regular follow-up care and supported by oncologists, with initial results showing a high participation rate, exceeding 90%, this bias should be minimal.

This multicenter cohort study prospectively investigates pulmonary dysfunction in young CCS. By assessing lung function as an intermediate outcome, rather than established disease or mortality, this study will provide a resource for evaluating pulmonary dysfunction at an earlier stage in the disease trajectory, particularly within the early years posttreatment. The initial response shows that integrating standardized pulmonary evaluations into routine follow-up care in Switzerland is feasible and widely accepted by both survivors and health care providers. The findings of this study will provide new insights to inform the development of guidelines and recommendations for pulmonary follow-up care.