Full Dose SBRT in Combination With Mediastinal Chemoradiation for Locally Advanced, Non-Small Cell Lung Cancer: A Practical Guide for Planning, Dosimetric Results From a Phase 2 Study, and a Treatment Planning Guide for the Phase 3 NRG Oncology LU-008 Trial.

PURPOSE: Stereotactic body radiation therapy (SBRT) has been used with high effectiveness in early-stage non-small cell lung cancer (NSCLC) but has not been studied extensively in locally advanced NSCLC. We conducted a phase 2 study delivering SBRT to the primary tumor followed by conventionally fractionated chemoradiation to the involved lymph nodes for patients with node-positive locally advanced NSCLC. This manuscript serves as both a guide to planning techniques used on this trial and the subsequent phase 3 study, NRG Oncology LU-008, and to report patient dosimetry and toxicity results. METHODS AND MATERIALS: We initiated a phase 2 multicenter single arm study evaluating SBRT to the primary tumor (50-54 Gy in 3-5 fractions) followed by conventionally fractionated chemoradiation to 60 Gy in 2 Gy fractions with doublet chemotherapy to the involved lymph nodes for patients with stage III or unresectable stage II NSCLC. Patients eligible for adjuvant immunotherapy received up to 12 months of durvalumab. We report a detailed guide for the entire treatment process from computed tomography simulation through treatment planning and delivery. The dosimetric outcomes from the 60 patients who completed therapy on study are reported both for target coverage and normal structure doses. We also report correlation between radiation-related toxicities and dosimetric parameters. RESULTS: Sixty patients were enrolled between 2017 and 2022. Planning techniques used were primarily volumetric modulated arc therapy for SBRT to the primary tumor and conventionally fractionated radiation to the involved nodes, with a minority of cases using dynamic conformal arc technique or static dynamic multileaf collimator intensity modulated radiation therapy. Grade 2 or higher pneumonitis was associated with lung dose V5 Gy > 70% and grade 2 or higher pulmonary toxicity was associated with lung dose V10 Gy > 50%. Only 3 patients (5%) experienced grade 3 or higher pneumonitis. Grade 2 or higher esophagitis was associated with esophageal doses, including mean dose > 20 Gy, V60 Gy > 7%, and D1cc > 55 Gy. Only 1 patient (1.7%) experienced grade 3 esophagitis. CONCLUSIONS: SBRT to the primary tumor followed by conventionally fractionated chemoradiation to the involved lymph nodes is feasible with planning techniques as described. Radiation-related toxicity on this phase 2 study was low. This manuscript serves as a guideline for the recently activated NRG Oncology LU-008 phase 3 trial evaluating this experimental regimen.

Automated, quantitative assessment of epidermal necrosis expression resulting from skin exposure to beta radiation.

PURPOSE: Radiation skin injuries are difficult to quantitatively assess. Various scoring scales exist based on visual images and can be used in quantitative form for histological scoring. As an alternative to human scoring systems, an automated, quantitative system is proposed to provide unbiased scoring of radiation skin injury biopsy samples, with comparisons to human-based scoring systems. MATERIALS AND METHODS: A unique algorithm was developed and tested on a sample pool obtained from in-vivo beta radiation experiments with a porcine model. The grading results achieved by the developed algorithm and those provided by an expert histopathologist are compared. RESULTS: The extent of the epidermal necrosis is quantified in terms of the number of dead cells and their respective distribution across the length of the samples. The accuracy of the grading performed by the automated algorithm is comparable to that of a trained histopathologist, as demonstrated by statistically significant difference between the grades. CONCLUSIONS: This study demonstrates the feasibility of the proposed method as a potential tool designed to aid in the histopathological analysis of the tissues affected by beta radiation exposure. An expanded study with a larger sample pool is recommended to further improve the accuracy of the proposed algorithm.