PubMed2001
PURPOSE: Cyclophosphamide is one of the most frequently used agents in the neoadjuvant, adjuvant, and high-dose chemotherapeutic treatment of breast cancers. Preclinical models indicate that cellular sensitivity to cyclophosphamide and other oxazaphosphorines, e.g., ifosfamide, is inversely related to the cellular content of two aldehyde dehydrogenases, viz ALDH1A1 and ALDH3A1, and glutathione. Breast tumor levels of these "determinants of cellular sensitivity to the oxazaphosphorines" are known to vary widely, and the decision as to whether or not to use an oxazaphosphorine as part of the therapeutic strategy to treat breast cancer in any given patient is likely to depend, in large part, on the levels of these determinants in that cancer. ALDH1A1, ALDH3A1, and glutathione levels can be easily quantified in primary breast tumors and in detectable metastatic breast tumors present in axillary lymph nodes because the amounts of tissue required for the desired analysis can be readily obtained, whereas these levels cannot be quantified in residual metastatic breast cancer cell populations, i.e., those that escape detection and/or that are inaccessible to surgical harvest. The inability to directly quantify residual metastatic breast cancer cell ALDH1A1, ALDH3A1, and glutathione levels would not preclude a rational decision with regard to the inclusion/exclusion of an oxazaphosphorine as part of the chemotherapeutic strategy intended to eradicate residual metastatic breast cancer cells if primary breast tumor levels of these determinants reliably predicted those in metastatic breast cancer cells. METHODS: ELISAs and spectrophotometric assays were used to quantify enzyme and glutathione levels in paired human primary and locally advanced metastatic breast tumor samples. RESULTS: Primary breast tumor ALDH1A1 and ALDH3A1 levels were highly predictive of their respective levels in paired metastatic breast tumors present in axillary lymph nodes (r2 = 0.80 and 0.85, respectively). On the other hand, those of glutathione were relatively poorly predictive of its levels in paired metastatic offshoots (r2 = 0.35). Primary breast tumor levels of some additional enzymes known to catalyze the detoxification/toxification of various anticancer agents, though not of cyclophosphamide, were poorly predictive (DT-diaphorase and glutathione S-transferases alpha, mu, and pi) or not predictive (cytochrome P450 1A1) of their respective levels in paired metastatic offshoots. CONCLUSION: Since ALDH1A1, ALDH3A1 and, to a lesser extent, glutathione levels in primary breast tumors reliably predicted those in detectable and easily accessible metastatic breast cancer cell populations, viz those in axillary lymph nodes, they are also likely to be predictive of these levels in undetectable and/or relatively inaccessible metastatic breast cancer cell populations. Thus, quantification of primary breast tumor ALDH1A1, ALDH3A1 and, to a lesser extent, glutathione levels prior to the initiation of not only neoadjuvant but also adjuvant and high-dose breast cancer chemotherapy is likely to be of value in the rational design of individualized chemotherapeutic regimens intended to eradicate breast cancer cells with a minimum of untoward effects.