Scientific basis for adjuvant and primary (neoadjuvant) chemotherapy.

It can be stated as a general biological principle that there are many compelling reasons why chemotherapy should be directed at minimal tumor burdens. This is true whatever the nature of the tumor and becomes especially valid when one is dealing with tumors that are not curable when treated at the advanced stage. The patients who are likely to have the greatest benefit from adjuvant chemotherapy are, somewhat paradoxically, those who are at the least risk for recurrence following primary treatment. This is because, on the average, these patients will have the least tumor burdens. Patients who are at very high risk for relapse in breast cancer, (stage II patients with four plus positive nodes) will be the ones with the greatest subclinical burdens and may well have already crossed the threshold of curability to incurability. Directing effective chemotherapy programs at patients with lesser risk of recurrence complicates the ethical problems associated with adjuvant chemotherapy. To some degree, these ethical concerns can be assuaged by the appreciation that it is likely that protracted programs of chemotherapy (1 to 2 years) may well not be necessary. In general, curative drug programs can generally accomplish objectives with 3 to 6 months of fairly intensive treatment. Reducing the duration of adjuvant cyclophosphamide, methotrexate, fluorouracil (CMF) from 12 months to 6 months did not appear to have an adverse effect on long-term results. Factors such as dose intensity and early use of effective noncross-resistant agents may be much more important than the chronic administration of agents in suboptimal dosage. The narrower question as to whether advancing the time forward of adjuvant chemotherapy will make additional significant impact on survival cannot be answered yet but clearly is an important issue. There are several theoretical reasons why neoadjuvant treatment might be of particular benefit, and even if it ultimately transpires that breast cancer is not an ideal model disease for this approach, it does not preclude this particular technique for being effective in other types of malignancy.