Cone-beam computed tomography in lung stereotactic ablative radiation therapy: predictive parameters of early...

To analyze lung lesion volume variations by contouring on cone-beam CT (CBCT) images to evaluate the early predictive parameters of stereotactic ablative radiation therapy (SABR) treatment response.

The prescribed dose of SABR was varied according to the tumour site (central or peripheral) and maximum diameter of the lesions by using a strategy of risk-adapted dose prescription with a dose range between 48 and 70 Gy in 3-10 consecutive fractions. For the purpose of the analysis, the gross tumour volume (GTV) was recontoured for each patient at first and last CBCT using two lung levels/windows: (a) -600/1000 HU and (b) -1000/250 HU. Univariate analysis was performed to evaluate a correlation between lung lesion variations on CBCT using the two levels/windows and treatment response 6 months after SABR. Independent variables were the number of fractions, time between initial and final fraction, biologically effective dose and pre-SABR GTV. Cut points of lesion volume reduction were evaluated to determine the correlation with complete response 6 months after SABR.

41 lung lesions were evaluated. 82 lung lesions were recontoured for each CBCT level/window. A lung lesion shrinkage of at least 20% was revealed to be statistically related to complete response 6 months after SABR for both the CBCT levels/windows used. The probability of complete response ranged between six and eight times higher in respect to CBCT levels/windows -600/1000 HU and -1000/250 HU, respectively, compared with patients without a lesion shrinkage of 20% at the last session of SABR.

According to current findings, a lung lesion shrinkage of at least 20% at the last session of SABR could be predictable of complete response 6 months thereafter. Further investigations about this topic are needed.

Prediction of the early tumour response could be useful to personalize imaging restaging after the completion of SABR or to incorporate additional therapies in case of poor responders to improve clinical outcomes.