Validation of the impact of heart base dose on survival in NSCLC patients from the PET-Plan Trial

Abstract

Purpose or Objective Heart dose has emerged as an independent predictor of overall survival in non-small cell lung cancer patients. Several studies have identified the base of the heart as the most significant region and a potential target for cardiac sparing. This work aims to further validate the impact of heart base dose in the multicentre, randomised PET-plan trial data and for the first time, will determine whether the effect remains significant when baseline cardiac function is included in the analysis. Material and Methods 205 patients with inoperable UICC stage II/III non-small cell lung cancer treated with 60-72 Gy in 2 Gy fractions in the PET Plan randomised controlled trial (NCT00697333) were included in this study. CT scans and dose distributions were spatially normalised to a reference patient using the “NiftyReg” deformable image registration package. For a given time point, mean dose distributions were calculated for the alive and dead patients, censored for follow-up. Dose differences between the groups were calculated and tested for significance by comparison to the null hypothesis dose difference distribution approximated by permutation testing. An anatomical region of interest was defined at 95% of the maximum t-value in the dataset and the mean dose to the region was extracted for all patients. A uni-variable analysis tested the association of survival with dose to the identified region, clinical variables including age, gender, performance status, tumour volume and baseline ejection fraction measured prior to radiotherapy. All variables were then included in a multivariable Cox proportional hazards model. Results 172 patients remained after processing and censoring for follow-up. At 2-years post treatment, a highly significant region was identified within the base of the heart (p < 0.005), centred on the origin of the left coronary artery and the region of the atrioventricular node (Figure 1). The median dose to the region was 33.5 Gy. Uni-variable analyses determined worse performance status, ejection fraction less than 50%, tumour volume and mean dose to the defined cardiac region to be significant. In multi-variable analysis, performance status (p = 0.03, HR: 5.69, 95% CI: 1.21 – 26.71), ejection fraction (p = 0.008, HR: 2.91, 95% CI: 1.32-6.40) and mean dose to the cardiac sub-region (p = 0.01, HR: 1.02, 95% CI: 1.00-1.03), remained significant (Table 1). Conclusion This work validates previous image-based data mining studies in identifying a cardiac region within the base of the heart as strongly associated with overall survival. Importantly, for the first-time baseline cardiac function was included with mean dose to the identified region in the multi-variable analysis. Poor ejection fraction did not negate the impact of dose to the base of the heart on survival.