|| Checking for direct PDF access through Ovid
A SERIES OF 1181 PATIENTS with carcinoma of the large bowel were entered into two prospective, randomized cooperative trials to study the possible effect of adjuvant 5-fluorouracil in conjunction with surgical resection. Drug was administered intravenously in standard 5-day courses, two courses being used in the first trial and sequential courses at 6-week intervals for an 18-month period in the Prolonged Intermittent Therapy (PIT) trial. On the basis of clinical and microscopic evidence of residual disease, patients were divided prior to assignment of therapy into “curative” and “palliative” groups. All eligible patients randomized to receive 5-fluorouracil are included in the treated group, even though 28 never received a single dose; others received less than the full treatment described in the protocol. Survival is significantly better (P = 0.0.5) for treated patients in the first trial and suggestive (P = 0.08) in the second (PIT) trial, using the Mantel-Haenszel chi-square test of the difference between two survival curves. The two trials actually represent live separate studies, and in all five instances survival in patients receiving adjuvant 5-fluorouracil exceeds that of patients treated by surgery alone. Although no single treatment difference is statistically significant, when viewed together the observed five treatment benefits in five comparisons are not likely to be due to chance (P < 0.01).