Evaluation of Observer Variation as Natural Experiment to Detect Sensitive Heart Subregions

PURPOSE/OBJECTIVE(S)

There is growing evidence associating dose to the base of the heart to reduced survival of lung cancer patients (McWilliam EJC 2017). However, randomized evidence on the benefits of sparing the base of the heart is missing. In this study we investigate variability in the shape between patients of heart contours used in planning as a natural experiment to evaluate selective heart sparing. The core assumption is that regions that are not included in the heart delineation will not be spared.

MATERIALS/METHODS

Data was collected for 1705 lung cancer patients treated in one center between 2010 and 2016 with IMRT/VMAT (55-66Gy in 20-33#) or SABR (54-60Gy in 3-8#), planned using manual heart contours, called delineation 1. Consistent reference heart contours were obtained using a commercial autocontouring software, called delineation 2. Heart shapes were mapped to spherical coordinates (φ, Φ, r) and the difference in radius Δr = r1 - r2 for each set of angles φ, Φ was calculated for all patients. A large Δr means that the manual delineation use for planning is relatively large in the direction. Cox-regression was performed for each set of angles using Δr, r2 and its interaction using overall survival as endpoint. Permutation testing is used to avoid multiple testing issues. The aim is to locate a region of the heart where bigger delineations lead to better sparing and hence better survival.

RESULTS

On average the heart base in our manual contours extends 34 mm more superior than automatic contours, because our protocol stipulates the inclusion of the full pericardial sac. Δr was not correlated with any clinical variables and is therefore a good candidate as instrumental variable in causal inference. Univariable Cox regression of Δr showed a uniform worsening of survival with larger manual delineations, likely due to reduced overall sparing given the use of volumetric dose constraints. After including the interaction with r2, no significant heart regions were found. However, analysis using the overall volume of the manual and auto-delineated heart did show a small but significant interaction effect where larger manual delineations improved outcome for smaller hearts. Our interpretation is that delineation variability relative to autocontouring (e.g., 1.9 mm SD at right atrium, up to 15 mm SD at apex) is not big enough to impact significantly on the heart dose and therewith survival because volumetric costs functions are used. In the future we will extend this analysis to include planned dose.

CONCLUSION

Variability in contouring in our cohort is not large enough to be used as a natural experiment to test the impact of selective heart sparing. However, larger volume delineations of small hearts are associated with reduced mortality, suggesting the importance of sparing the base of the heart where most contouring variability occurs.

Effect of systemic inflammation biomarkers on overall survival after lung cancer radiotherapy: a single-center large-cohort study

INTRODUCTION

Recent studies suggest that immune-related cells can be recruited for anti-tumor functions as well as tumor progression and the interplay between systemic inflammation and local immune response may play a major role in the development and progression of various cancers including lung cancer. Inflammatory markers, such as neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and systemic immune-inflammation index (SII) can be used as surrogate biomarkers of host immune status. In this work, associations between neutrophils, lymphocytes, platelets, NLR, PLR, SII and overall survival (OS) are investigated in two cohorts of non-small cell lung cancer (NSCLC) patients treated with fractionated radiotherapy (RT) and stereotactic body radiation therapy (SBRT) and a cohort of small cell lung cancer (SCLC) patients treated with fractionated RT.

MATERIAL AND METHODS

Data from 2513 lung cancer patients were retrospectively analyzed. Baseline NLR, PLR, and SII (NLR × platelet count) were calculated from full blood test prior to RT initiation. Cox proportional hazards regression analyses were used to evaluate the association between systemic inflammation markers and known clinical factors with OS.

RESULTS

The two-year OS was 42%, 63%, and 62% in the NSCLC fractionated RT, SBRT, and SCLC cohort. NLR (per 1 unit: hazard ratio [HR]: 1.04, p < 0.05) and SII (per 100 × 109/L: HR: 1.01, p < 0.05) remained the strongest independent factors of OS in multivariable Cox analyses, correcting for clinical factors in early-stage and locally advanced NSCLC and SCLC patients treated with RT.

DISCUSSION

This single-center large-cohort study suggests that baseline NLR and SII are independent prognostic biomarkers associated with OS in locally advanced and early-stage NSCLC patients treated with either curative-intent fractionated RT or SBRT and SCLC patients treated with curative-intent fractionated RT. External validation is warranted to evaluate the utility of these biomarkers for patients' stratification and adapting new treatment approaches.

A Predictive Model for Reactive Tube Feeding in Head and Neck Cancer Patients Undergoing Definitive (Chemo) Radiotherapy

AIMS

Careful management of a patient's nutritional status during and after treatment for head and neck squamous cell cancers (HNSCC) is crucial for optimal outcomes. The aim of this study was to develop a model for stratifying a patient's risk of requiring reactive enteral feeding through a nasogastric tube during radiotherapy for HNSCC, based on clinical and treatment-related factors.

MATERIALS AND METHODS

A cohort of consecutive patients treated with definitive (chemo)radiotherapy for HNSCC between January 2016 and January 2018 was identified in the institutional electronic database for retrospective analysis. Patients requiring enteral feeding pretreatment were excluded. Clinical and treatment data were obtained from prospectively recorded electronic clinical notes and planning software.

RESULTS

Baseline patient characteristics and tumour-related parameters were captured for 225 patients. Based on the results of the univariate analysis and using a stepwise backwards selection process, clinical and dosimetric variables were selected to optimise a clinically predictive multivariate model, fitted using logistic regression. The parameters found to affect the probability, P, of requiring a nasogastric feeding tube for >4 weeks in our clinical multivariate model were: tumour site, tumour stage (early T0/1/2 stage versus advanced T3/T4 stage), chemotherapy drug (none versus any drug) and mean dose to the contralateral parotid gland. A scoring model using the regression coefficients of the selected variables in the clinical multivariate model achieved an area under the curve (AUC) of 0.745 (95% confidence interval 0.678-0.812), indicating good discriminative performance. Internal validation of the model involved splitting the dataset 80:20 into training and test datasets 10 times and assessing differences in AUC of the model fitted to these.

CONCLUSIONS

We developed an easy-to-use prediction model based on both clinical and dosimetric parameters, which, once externally validated, can lead to more personalised treatment planning and inform clinical decision-making on the appropriateness of prophylactic versus reactive enteral feeding.

Skin Toxicity Profile of Photon Radiotherapy versus Proton Beam Therapy in Paediatric and Young Adult Patients with Sarcomas

AIMS

Radiotherapy is key in the management of patients with both Ewing sarcoma and rhabdomyosarcoma. However, there is little evidence in the literature with regards to radiation-induced skin toxicities (RISTs) for patients treated with conventional radiotherapy with X-rays (XRT) or proton beam therapy (PBT) for these two conditions. In the present study we evaluated acute and late RIST in patients treated within European protocols with either PBT or XRT, taking both clinical and dosimetric variables into consideration.

MATERIALS AND METHODS

This was a retrospective analysis of 79 paediatric/young adult patients treated with radical radiotherapy (with XRT or PBT) and concurrent chemotherapy. In all cases, radiotherapy was given in conventional fractionation (1.8 Gy/fraction). Acute and late RISTs were registered according to the Radiation Therapy Oncology Group (RTOG) scoring system.

RESULTS

With regards to acute RIST, 47.9% (23/48) of XRT patients and 48.4% (15/31) of PBT patients had acute grade 2/3 toxicity. When it comes to late RIST, 17.5% (7/40 with known toxicity profile) of XRT patients and 29.0% (9/31) of PBT patients had grade 1/2 toxicity. This difference of -11.5% (95% confidence interval -31.2 to 7.9%) in grade 1/2 toxicity between XRT and PBT was not statistically significant (P = 0.25). Regardless of the radiotherapy technique, V30Gy seems a good predictor of acute RIST. Moreover, for the same value of V30Gy, patients who receive PBT may have a higher risk of moderate-severe acute RIST. Perhaps due to the small sample, definitive conclusions on the predictive factors of late RIST could not be drawn.

CONCLUSIONS

No clinically meaningful differences in acute and late RIST were observed between PBT and XRT subgroups. Systematic differences in the modelling of the build-up region may exist between XRT and PBT algorithms, which could make the comparison of dose metrics between techniques potentially biased. A more comprehensive analysis of dosimetric data on larger patient cohorts is needed to elucidate the most relevant skin dose metrics. Dose-effect models of RIST for this unique patient population would be an invaluable tool in radiotherapy plan optimisation.

Cardiovascular mortality and morbidity following radical radiotherapy for lung cancer: Is cardiovascular death under-reported?

BACKGROUND

Lung cancer is the most common malignancy worldwide. Radical radiotherapy is an essential treatment in the management of early and locally advanced lung cancer. Cardiac events are known to occur following radical radiotherapy for lung cancer. This study examines the burden of cardiac events post radiotherapy, and estimates the accuracy of death certification in patients who received radical radiotherapy for lung cancer.

METHODS

We conducted a retrospective observational cohort study for all patients receiving radical radiotherapy for non-small cell lung cancer (NSCLC) at a large cancer centre between 01/01/2010 to 31/12/2016. Baseline cardiovascular disease and cancer status and treatment data were collected, along with hospital admission data and documented cause of death from the national registry for a median follow-up period of 34 months.

RESULTS

Of 1224 patients included in the analysis, 378 (30.9%) patients had cardiovascular disease at baseline, including 140 (11.4%) with prior myocardial infarction. In the 846 patients without known cardiovascular disease, 451 (53.3%) had a QRISK2 predicted 10-year cardiovascular risk >20% over 10 years. During follow-up, 215 hospitalisations occurred (Incidence rate 6.2 per hundred patient years) which were classified as primarily cardiac, and 622 patients died (18 per 100 patient-years). However, death certificates stated a primary cardiac cause of death in only 33 cases (5.3% of deaths). Notably, 29% of patients dying out of hospital and certified as cancer death did not have documented cancer relapse prior to death, and 61% had no community palliative care input prior to death, implying these events may have been sudden and unexpected.

CONCLUSION

There is a high prevalence of baseline cardiovascular disease in people undergoing radiotherapy for NSCLC, accompanied by significant rates of post-radiotherapy cardiovascular hospitalisation. However, only a small proportion of deaths are attributed to cardiovascular disease, together with the large amount of sudden deaths observed, this suggests that cardiovascular death is greatly under-reported in official statistics.