Individually Tailored Toxicity-based Chemotherapy : Studies on Patients with Primary and Metastatic Breast Cancer

Standard dosing of chemotherapy based on body surface area (BSA) results in large individual differences in toxicity due to a large inter-patient variability in pharmacokinetics (PK) and pharmacodynamics (PD). This results in under-dosing in certain patients with a potentially weaker antitumoral effect. Three clinical studies of individually tailored dosing of chemotherapy, based on haematological toxicity were conducted. In the first study, 26 women with metastatic breast cancer were treated with tailored and dose-escalated 5-fluorouracil, epirubicin and cyclophosphamide, supported by G-CSF (dFEC). In the second study 525 patients with high-risk primary breast cancer were randomised between dFEC and high-dose chemotherapy with autologous bone-marrow transplantation. The feasibility of a FEC regimen with doubled cyclophosphamide dose to mobilise peripheral stem cells was investigated. In the third study, 44 metastatic patients were treated with tailored epirubicin and docetaxel (ET). PK and PD were also investigated in these patients. The potential effects of G-CSF on MRI tumour evaluation were studied in 18 patients with skeletal metastases. Toxicity-based dosing entailed an evenly distributed two- to three-fold range of tolerated doses in all three studies. Efficacy and toxicity were not correlated to tolerated dose-levels. Tailored dFEC resulted in a response rate of 81% and the same regimen resulted in fewer breast cancer relapses compared with standard FEC followed by high-dose therapy. Toxicity was manageable except for an increased rate of secondary leukaemia. The modified FEC could safely mobilise sufficient numbers of stem-cells. Tailored ET resulted in a response rate of 63%. The inter-individual variability in drug clearance was larger than the inter-occasion variability and a semi-physiological model of PK and PD could predict leukocyte nadir and duration. An increased diffuse MR signal in the long TE IR-TSE sequence was observed in normal bone-marrow during G-CSF treatment; this could be mistaken as disseminated metastatic disease and could obscure focal metastases. In conclusion, the concept of individually tailored toxicity-based dosage of chemotherapy was equally feasible in primary and metastatic breast cancer, in two different chemotherapy regimens and in treatment with or without G-CSF support and may provide a pragmatic way of overcoming the shortcomings of standard BSA-based dosing