A novel and objective plan evaluation tool for dose escalation in NSCLC within the ADCSCaN trial

Abstract

Purpose or Objective ADSCaN (Accelerated, Dose-escalated, Sequential Chemo- Radiotherapy in Non-small cell lung Cancer (NSCLC); ISRCTN47674500) is a randomised phase II clinical trial comparing four experimental radiotherapy (RT) regimes for NSCLC (CHART-ED, IDEAL, I-START and ISOTOXIC IMRT) to a standard RT regime of 55Gy in 20 fractions. RT planning for IDEAL, I-START and Isotoxic IMRT includes novel dose escalation and intensification schemes which require a change in planning technique. The aim of this study was to identify and to validate novel metrics to assess whether an RT plan is optimal. Material and Methods A NSCLC benchmark case was sent to ADSCaN centres. Centres submitted a plan for at least one of the experimental arms. GTV, CTV and OARs were precontoured. The PTV was derived at each centre as per the protocol; 5mm isotropic margin was used by centres performing per fraction cone-beam CT, and 7mm axially and 9mm cranio-caudally for those who did not. Table 1 shows the ADSCaN research arm requirements for RT planning. PTV conformity (V95%isodose /PTV Vtotal) and OAR compromise (OAR V95%isodose/OAR&PTV VTotal overlap) indices were calculated and validated for each dose escalation arm from the 22 submitted plans. This analysis focuses on the 3 research arms that employ isotoxic dose escalation methods; therefore CHART-ED arm was not included in this analysis. Results The OAR limiting dose escalation for IDEAL and I-START plans was the oesophagus, whereas the mediastinal envelope limits prescription dose for ISOTOXIC IMRT. Table 2 shows the dose achieved to PTV, oesophagus, mediastinal envelope, PTV conformity and OAR compromise indices. The remaining OAR doses although assessed, are not reported in this table as they were easily achieved for this case. All plans were assessed as per the requirements described (table 2) and were considered suboptimal if not achieving the criteria for at least two of the following: PTV coverage, OAR dose-volume constraints, PTV conformity and OAR compromise indices. All plans were also visually assessed in all slices. Conclusion 45.5% of the submitted plans were sub-optimal as plans had compromised PTV coverage in favour of meeting OAR dose constraints, overdosed OARs or did not escalated the prescription dose as high is achievable. We attribute the high number of sub-optimal plans to the learning curve in the new individualised planning techniques included in ADSCaN. Assessing plans that use novel dose escalation can be challenging, particularly in determining whether a plan is optimal. This is a novel and objective evaluation that can be applicable to any dose escalated lung RT plan.