Adaptive intrapatient dose escalation of cisplatin in patients with advanced head and neck cancer.

The purpose of this study was to explore the feasibility and toxicity of intrapatient dose adjustment using predefined levels of exposure to cisplatin, with the ultimate goal to further improve the antitumor activity of the treatment. The primary parameter for adaptive dosing was the level of platinum DNA adducts in peripheral white blood cells (WBC) and the secondary parameter the area under the curve (AUC) of unbound platinum in plasma, which were determined during the applied courses. Target levels had been defined in a previously performed pharmacologic study. The concept of adaptive dosing was tested in 16 patients with locally advanced head and neck (H/N) cancer who would receive six weekly courses of cisplatin at a starting dose level of 80 mg/m(2), which was previously investigated in a phase II study. Forty-seven percent of patients received a dose increase varying from 10 to 40%. Only two patients had exposure levels significantly below the defined target levels for DNA adducts and AUC. The majority of patients reached the defined target levels by modest dose increases of 10-20% during course 2. Relevant but reversible ototoxicity (temporary grade 3 in two patients) and renal toxicity (temporary grade 2 in two other patients) were observed. The pattern and severity of the toxicity was comparable to that encountered in the previous phase II study in H/N cancer patients. We conclude that the strategy of intrapatient dose adjustment for cisplatin is practically feasible in a research setting even when a short turn around time of 1 week is the limit for reporting results. Although in some patients the dose increase that had to be applied to reach target levels was substantial (up to 40%), this approach in H/N cancer patients is not expected to improve the response rate significantly, because these significantly underdosed patients represented only a small percentage of the investigated population. The great majority of patients needed only limited (10-20%) dose increases which very likely will not improve the response rate to a clinically significant extent. The outlined concept is currently being explored in other tumor types and schedules of cisplatin.