Purpose: We aimed to report the final toxicity results on a radiation-dose escalation trial designed to test a hypothesis that very high doses of radiation could be safely administered to patients with non-small-cell lung cancer (NSCLC) by quantifying the dose-volume toxicity relationship of the lung.
Methods and materials: A total of 109 patients with unresectable or medically inoperable NSCLC were enrolled and treated with radiation-dose escalation (on the basis of predicted normal-lung toxicity) either alone or with neoadjuvant chemotherapy by use of 3D conformal techniques. Eighty-four patients (77%) received more than 69 Gy, the trial was stopped after the dose reached 103 Gy. Estimated median follow-up was 110 months.
Results: There were 17 (14.6%) Grade 2 to 3 pneumonitis and 15 (13.8%) Grade 2 to 3 fibrosis and no Grade 4 to 5 lung toxicity. Multivariate analyses showed them to be (1) not associated with the dose prescribed to the tumor, and (2) significantly (p<0.001) associated with lung-dosimetric parameters such as the mean lung dose (MLD), volume of lung that received at least 20 Gy (V20), and the normal-tissue complication probability (NTCP) of the lung. If cutoffs are 30% for V20, 20 Gy for MLD, and 10% for NTCP, these factors have positive predictive values of 50% to 71% and negative predictive value of 85% to 89%.
Conclusions: With long-term follow-up for toxicity, we have demonstrated that much higher doses of radiation than are traditionally administered can be safely delivered to a majority of patients with NSCLC. Quantitative lung dose-volume toxicity-based dose escalation can form the basis for individualized high-dose radiation treatment to maximize the therapeutic ratio in these patients.