The future for oral cancer chemotherapy has never been brighter. The need to move cancer treatment from a predominantly hospital-based, inpatient system into the ambulatory setting has joined with the growing body of information demonstrating higher antitumor activity, lower systemic toxicity, or both with dosing regimens that produce prolonged exposure to some cancer chemotherapy. This has led to further exploration of oral administration of anticancer drugs that have been available for many years (ie, etoposide, cyclophosphamide, and idarubicin) and novel strategies for oral use of anticancer drugs traditionally administered by the iv route (paclitaxel plus cyclosporin, 5-flurouracil plus eniluracil), leading to a new era in the administration of cancer chemotherapy (1). The promise for oral chemotherapy is well illustrated by the use of mercaptopurine in maintenance therapy for childhood acute lymphoblastic leukemia. The daily administration of oral mercaptopurine during many weeks of continuation therapy is an important component of most treatment protocols for childhood acute lymphoblastic leukemia (2, 3), and this schedule could not be conveniently achieved with iv therapy. Oral cyclophosphamide has been an important component of adjuvant therapy for breast cancer for over a decade, permitting self administration in a convenient setting and allowing patients to have a greater role in their therapy. With the development of oral anthracyclines (such as idarubicin) and less variable approaches for oral 5-fluorouracil administration (such as coadministration with eniluracil), treatment regimens with only oral chemotherapy are now under clinical evaluation for adjuvant breast cancer (4). The significant schedule dependency of etoposide, 5-fluorouracil, topoisomerase I inhibitors, and other classes of chemotherapy represent areas in which chronic oral administration of chemotherapy may make a significant difference for patients with cancer.

Many of the pitfalls of oral chemotherapy can be anticipated from well-documented experiences with other therapeutic agents, including variable absorption, unpredictable and incomplete bioavailability, and uncertainty about patient compliance. Nearly all medications demonstrate a high degree of variation in oral bioavailability among patients. The report by Hande et al.(5) in this issue of Clinical Cancer Research highlights this problem. Inter-and intrapatient variability in etoposide area under the curve were evaluated using an elegant stable isotope dilution method that allowed simultaneous administration of iv and oral medication. The investigators were able to demonstrate that intrapatient variation in iv etoposide (10%), was much less than the 22.2% intrapatient coefficient of variation observed after oral etoposide administration. The large intrapatient variability observed with oral administration is similar to that seen with other medications that undergo a significant amount of metabolism by small bowel and/or hepatic P-450 enzymes. Not surprisingly, a greater degree of interpatient variability (2–3-fold), compared with intrapatient variability, was observed after both iv and oral administration of etoposide. The report makes two points with broad implications for the development of oral chemotherapy:(a) The administration of repeated doses of a drug to the same patient will have less pharmacokinetic variation than that observed in the overall patient population. This is not unanticipated, because genetic sources of variability are eliminated when comparisons are made within the same individual. The lower intrapatient variability suggests that approaches such as individualized therapy based on measurement of …