[Effect of timing of granulocyte-colony stimulating factor administration on leukopenia induced by systemic chemotherapy in patients with non-small-cell lung cancer--multi-center randomized crossover study]

Sixty-six chemotherapy-naive patients with non-small-cell carcinoma of the lung were given two courses of systemic chemotherapy consisting of mitomycin C, vindesine, and cisplatin. The effect of the timing of administration of grannulocyte colony-stimulating factor (G-CSF) on the incidence of neutropenic fever, the nadir leukocyte count, the duration of neutropenia ( < or = 1000/mm3), and the time needed for recovery from neutropenia was studied. Patients were assigned at random to begin receiving G-CSF (50 microns/m2, subcutaneously) either when the leukocyte count was less than or equal to 1000/mm3 (group I) or when it was between 1000/mm3 and 2000/mm3 (group II), in a crossover fashion. The nadir leukocyte count was lower in group I than in group II (859/mm3 and 1215/mm3, respectively). The duration of leukopenia (defined as a leukocyte count less than or equal to 1000/mm3) was greater in group I than in group II (1.5 days and 0.8 day, respectively), as was the time needed for recovery to a leukocyte count of 2000/mm3 (1.9 days and 1.6 days, respectively) (p < 0.05). No differences were found in the incidence of neutropenic fever (group I: 44%, group Ii: 45%), in the duration of fever (group I: 2.3 days, group II: 2.8 days), or in the duration of G-CSF use (group I: 6.3 days, group II: 6.8 days). There were no treatment-related deaths in either group. We conclude that when this type of combination chemotherapy is given for non-small-cell carcinoma of the lung, administration of G-CSF can be postponed without clinical problems until the leukocyte count is less than of equal to 1000/mm3.