Special problems in the evaluation of results in adjuvant trials of cancer treatment

Publisher Website
Google Scholar
Abstract
Adjuvant trials in cancer treatment present special problems in statistical analysis. When the primary treatment modality results in a relatively high cure rate and additional methods of treatment are being inserted in an effort to raise the figures to even higher levels, therapeutic benefit can be obscured unless all features of the study are carefully considered. Biologic benefit may easily be masked by the sizable number of patients randomized to receive adjuvant treatment who have no residual tumor and therefore could not possibly benefit therefrom. For example, 50% of patients undergoing “curative” resection for colon cancer survive without evidence of disease beyond the five‐year postoperative period; thus at least half the patients randomized to receive adjuvant treatment (chemotherapy, radiotherapy, immunotherapy) will be disease‐free at the time of randomization and cannot benefit from adjuvant therapy. New methods of assessment of treatment must be developed; otherwise, substantial therapeutic effectiveness may go undetected.