Abstract
BACKGROUND AND PURPOSE: Loco-regional recurrence is seen in up to 50 % of patients with non-small cell lung cancer (NSCLC) by 2 years after curative intent radiotherapy (RT). For these patients, radical re-irradiation (re-RT) is possible, but data is lacking regarding efficacy and radiation dose response, especially about conventionally fractionated re-irradiation. We analysed associations between survival following re-irradiation and RT dose, to guide clinicians regarding target re-RT dose, and predict re-treatment efficacy. MATERIAL AND METHODS: We performed a literature search for studies primarily comprised of NSCLC patients that detailed 2-year overall survival (OS(2-yr)) rates and delivered doses. These data were collated with intervals between treatments, PTV sizes and use of concurrent chemotherapy where this information was available. Logistic regression analyses of associations between OS(2-yr) and treatment and patient factors were carried out. Doses required for 30% and 50% OS(2-yr) were calculated. RESULTS: We identified 20 suitable studies (675 patients). In univariable models, OS(2-yr) was significantly associated with the initial RT dose, re-RT dose and chemotherapy use but not the interval. The best multivariable OS(2-yr) model according to the Akaike Information Criterion included only the re-RT dose (p < 0.05) and described the data well (Hosmer-Lemeshow p-value = 0.385). This model predicted OS(2-yr) rates of 30 % and 50 % at re-RT equivalent doses in 2 Gy fractions (EQD2s) of 49.8 Gy(10) (95 % CI 36.4, 58.0 Gy(10)) and 76.5 Gy(10) (95 % CI 70.8, 82.7 Gy(10)) respectively. CONCLUSION: OS(2-yr) following re-RT of recurrent NSCLC is significantly associated with retreatment dose. A reasonable target dose for re-RT is EQD2s > 50 Gy(10) with survival rates continuing to increase to 85 Gy(10).