Ej. Anaissie et al., INFECTIONS IN PATIENTS WITH CHRONIC LYMPHOCYTIC-LEUKEMIA TREATED WITHFLUDARABINE, Annals of internal medicine, 129(7), 1998, pp. 559-566
Background: Fludarabine, a purine analogue with activity in chronic ly
mphocytic leukemia, is usually well tolerated. Although serious infect
ions after fludarabine therapy have been described, a systematic analy
sis of the risk factors for such infections in chronic lymphocytic leu
kemia is lacking. Objective: To determine the risk factors for major i
nfection in patients with chronic lymphocytic leukemia treated with fl
udarabine. Design: Retrospective review of medical records. Setting: C
ancer center. Patients: 402 patients with chronic lymphocytic leukemia
not previously treated or treated with chlorambucil (with or without
prednisone) who received fludarabine (30 mg/m(2) of body surface area
per day for 5 days) with or without prednisone at 4-week intervals. Re
sults: Infections occurred more often in previously treated (144 of 24
8 [58%]) than in previously untreated (53 of 154 [34%]) patients (P <
0.001). Listeriosis or pneumocystosis occurred in 12 of 170 (7%) previ
ously treated patients receiving fludarabine plus prednisone, 0 of 78
previously treated patients receiving fludarabine alone, and 2 of 154
(1%) previously untreated patients receiving fludarabine plus predniso
ne (P = 0.003). Univariate analysis identified previous chemotherapy,
advanced disease, failure to respond to fludarabine, elevated serum P,
microglobulin level (P < 0.001), low serum albumin level (P = 0.024),
elevated serum creatinine concentration (P = 0.008), and low granuloc
yte count (P = 0.003) as risk factors for infection. Multivariate anal
ysis identified Rai stage III or IV (odds ratio, 1.98 [95% CI, 1.17 to
3.94]), previous treatment (odds ratio, 2.24 [CI, 1.43 to 3.51]), and
elevated serum creatinine concentration (odds ratio, 1.98 [CI, 1.09 t
o 3.67]) as statistically significant independent risk factors for maj
or infection. A baseline granulocyte count of more than 1000 cells/mu
L was protective (odds ratio, 0.54 [CI, 0.29 to 0.99]). Five (26%) of
19 patients with a CD4 count less than 50 cells/mL developed cutaneous
tester compared with 9 (6%) of 139 patients with a CD4 count greater
than 50 cells/mL (P = 0.01). Conclusions: Fludarabine used in previous
ly treated patients with chronic lymphocytic leukemia may be associate
d with infections involving T-cell dysfunction, such as listeriosis, p
neumocystosis, mycobacterial infections, and opportunistic fungal and
viral infections. Prophylaxis or presumptive therapy should be initiat
ed in the appropriate setting.