Machine-learning-based prediction of the effectiveness of the delivered dose by exhale-gated radiotherapy for locally advanced lung cancer: The additional value of geometric over dosimetric parameters alone

Abstract

Purpose This study aimed to assess interfraction stability of the delivered dose distribution by exhale-gated volumetric modulated arc therapy (VMAT) or intensity-modulated arc therapy (IMAT) for lung cancer and to determine dominant prognostic dosimetric and geometric factors. Methods Clinical target volume (CTV Plan ) from the planning CT was deformed to the exhale-gated daily CBCT scans to determine CTV i , treated by the respective dose fraction. The equivalent uniform dose of the CTV i was determined by the power law ( g EUD i ) and cell survival model (EUD iSF ) as effectiveness measure for the delivered dose distribution. The following prognostic factors were analyzed: (I) minimum dose within the CTV i (D min_i ), (II) Hausdorff distance (HDD i ) between CTV i and CTV Plan , (III) doses and deformations at the point in CTV Plan at which the global minimum dose over all fractions per patient occurs (PD min_global_i ), and (IV) deformations at the point over all CTV i margins per patient with the largest Hausdorff distance (HDPw orst ). Prognostic value and generalizability of the prognostic factors were examined using cross-validated random forest or multilayer perceptron neural network (MLP) classifiers. Dose accumulation was performed using back deformation of the dose distribution from CTV i to CTV Plan . Results Altogether, 218 dose fractions (10 patients) were evaluated. There was a significant interpatient heterogeneity between the distributions of the normalized g EUD i values ( p <0.0001, Kruskal–Wallis tests). Accumulated g EUD over all fractions per patient was 1.004–1.023 times of the prescribed dose. Accumulation led to tolerance of ~20% of fractions with g EUD i < 93% of the prescribed dose. Normalized D min >60% was associated with predicted g EUD values above 95%. D min had the highest importance for predicting the g EUD over all analyzed prognostic parameters by out-of-bag loss reduction using the random forest procedure. Cross-validated random forest classifier based on D min as the sole input had the largest Pearson correlation coefficient (R=0.897) in comparison to classifiers using additional input variables. The neural network performed better than the random forest classifier, and the g EUD values predicted by the MLP classifier with D min as the sole input were correlated with the g EUD values characterized by R=0.933 (95% CI, 0.913–0.948). The performance of the full MLP model with all geometric input parameters was slightly better (R=0.952) than that based on D min ( p =0.0034, Z-test). Conclusion Accumulated dose distributions over the treatment series were robust against interfraction CTV deformations using exhale gating and online image guidance. D min was the most important parameter for g EUD prediction for a single fraction. All other parameters did not lead to a markedly improved generalizable prediction. Dosimetric information, especially location and value of D min within the CTV i , are vital information for image-guided radiation treatment.

Topics & Concepts