Background and purpose: Radiotherapy is a treatment method frequently employed in the management of thoracic tumours. Although the highest incidence of these tumours is found in elderly people, tolerance to radiotherapy is not well documented in older age groups. Many physicians are tempted to alter the radiotherapy planning in a population with a supposed lower life expectancy in order to prevent acute reactions whereas late reactions are often ignored. The current study aimed to determine the influence of age on the frequency and severity of acute and late side-effects and also whether the prognosis of tumours sufficiently differed between ages to justify different attitudes towards their management.
Materials and methods: Data from 1208 patients receiving chest irradiation and included in arms designed with RT of six EORTC randomized trials were evaluated. Data were extracted by a computer program elaborated for each study and were merged in a single database for analysis. Patients were split into six age ranges from 50 to 70 years and over. Survival and late toxicity were calculated with the Kaplan-Meier method and comparison between age groups was performed with the logrank test. The gamma-statistic test was used to test the impact of age on acute toxicity occurrence.
Results: Survival adjusted for the primary location of the tumour was comparable in each age group (P = 0.82). Data regarding age and acute toxicity were available for 1208 patients who experienced 640 grade > or =1 toxicities. The difference in distribution over age was not significant for acute nausea, dyspnea, oesophagitis, weakness and WHO performance status alteration. Weight loss was significantly different with regards to age with a trend toward increased weight loss in older age groups (P = 0.002). To minimize actuarial bias, only patients surviving more than 90 days were analyzed for late effect risks. Late toxicities were examined only if they occurred before an eventual tumour failure in order to avoid confusion between effects of first and second line treatments. In such conditions, 1082 grade > or =1 late toxicities were recorded in 935 patients of 1106 available for analysis. The mean time to complication was 13 months and was similar in all age groups. Forty percent of patients were free of complication at 4 years, the logrank test showing no significant difference between age groups (P = 0.57). For grade >2 side-effects, the calculation did not show any difference between each age group (P = 0.1). A detailed analysis of late dyspnea and late weakness studied with the same method did not demonstrate any difference between age groups. Only grade >2 late oesophagitis demonstrated a significant trend to be more frequent in older patients (P = 0.01), but this difference disappeared after adjustment on study (P = 0.32).
Conclusion: The absence of toxicity observed in the current study regardless of age reinforces the conviction that age per se is not a sufficient reason to exclude patients in good general condition with thoracic tumour from curative radiotherapy when medically indicated.