A simple measure of reliability in clinical trials of surgical adjuvant radiotherapy

Aims The ‘number needed to treat’ (NNT) is a measure of therapeutic effect and is defined as the number of patients needed to be treated, so that the treatment of one of them is successful. Since many patients or physicians find this way to describe therapeutic potency very intuitive, I calculated the NNTs for the surgical adjuvant radiotherapy of different malignant neoplasias. Methods This computation was applied to selected, prospective, randomized trials or meta-analyses on adjuvant or adjunctive radiotherapy of epithelial or glial malignancies. Results Among 17 NNTs calculated for trials reporting improvements in local relapse rates 2 are smaller than 5, 8 are in the range from 5 to 10 and 7 are larger than 10 (median follow-up 5 years; range 3–20). In contrast, only 2 NNTs from 10 trials reporting survival advantages are smaller than 10 (median follow-up 4 years; range 1–20). Conclusions NNTs correctly reflect that adjuvant radiotherapy has a far more pronounced impact on the probability of local recurrence of a malignancy as opposed to patientʹs survival.