291 Failure of Fluconazole Prophylaxis to Reduce Mortality or the Requirement of Systemic Amphotericin B Therapy during Treatment for Refractory Acute Myeloid Leukemia Results of a Prospective Randomized Phase III Study Wolfgang Kern, M.D.1 Gerhard Behre, M.D.1 Thomas Rudolf, M.D.1 Andrea Kerkhoff, M.D.2 Albert Grote-Metke, M.D.3 Hartmut Eimermacher, M.D.4 Ursula Kubica, M.D.5 Bernhard Wörmann, M.D., Ph.D.1 Thomas Büchner, M.D., Ph.D.2 Wolfgang Hiddemann, M.D., Ph.D.1 for the German AML Cooperative Group 1 Department of Hematology and Oncology, GeorgAugust-University, Göttingen, Germany. 2 Department of Hematology and Oncology, Westfälische Wilhelms University, Münster, Germany. 3 Evangelisches Krankenhaus, Hamm, Germany. 4 Katholisches Krankenhaus Hagen, Hagen, Germany. 5 Zentralkrankenhaus St. Jürgen, Bremen, Germany. Presented at the Annual Congress of the German Society for Hematology and Oncology, Düsseldorf, Germany, 1996 (Kern W, Behre G, Büchner Th, Hiddemann W, for the German AML Cooperative Group. Failure of fluconazole prophylaxis to reduce mortality during treatment for refractory acute myeloid leukemia: a phase III multicenter study. Ann Hematol 1996;73[Suppl 2]:A10). The following institutions and investigators participated in the study (in order of the number of enrolled patients): University of Münster (Büchner T, Kerkhoff A); University of Göttingen (Hiddemann W, Wörmann B, Kern W); Evangelisches Krankenhaus, Hamm (Grote-Metke A); St. Martin Hospital, Hagen (Eimermacher H); Zentralkrankenhaus St. Jürgen, Bremen (Kubica U); Städtisches Krankenhaus Süd, Lübeck (Bartels H); Evangelisches Kranken© 1998 American Cancer Society BACKGROUND. Invasive fungal infections have increasingly become a matter of concern with regard to patients receiving intensive myelosuppressive therapy for hematologic malignancies. Such infections, especially prolonged neutropenia systemic fungal infections, may contribute substantially to infectious complications and early death. Measures for early detection and effective prophylactic strategies using active and nontoxic antifungal agents are therefore urgently needed. METHODS. The current randomized study was initiated to assess the efficacy of oral fluconazole as systemic antifungal prophylaxis for high risk patients with recurrent acute myeloid leukemia undergoing intensive salvage therapy. RESULTS. Of 68 fully evaluable patients, 36 were randomized to fluconazole in addition to standard prophylaxis with oral co-trimoxazol, colistin sulphate, and amphotericin B suspension, and 32 were randomized to standard prophylaxis only. No major differences between the two groups were observed in the number of episodes of fever of unknown origin (61% vs. 50%) or clinically defined infections (56% vs. 50%). Microbiologically defined infections were more frequent in the fluconazole group (50% vs. 31%), mainly due to a higher incidence of bacteremias (42% vs. 22%). There were two cases of proven invasive fungal infections in each group. Systemic amphotericin B was administered more frequently to patients receiving fluconazole prophylaxis (56% vs. 28%). Fluconazole prophylaxis had no impact on the rate of early death or overall survival. CONCLUSIONS. For patients with high risk recurrent acute myeloid leukemia undergoing intensive salvage therapy, antifungal prophylaxis with fluconazole was not superior to standard prophylaxis only. Cancer 1998;83:291–301. © 1998 American Cancer Society. KEYWORDS: fluconazole, fungal infections, prophylaxis, acute myeloid leukemia. haus, Essen (Tirier C); Allgemeines Krankenhaus Altona, Hamburg (von Seydewitz T); University of Kiel (Gassmann W, Zurborn KH); University of München (Emmerich B, Hanauske AR); Krankenhaus Düren (Karow J); St. Johannes Hospital, Duisburg (Schadek-Gressel B); University of Hamburg (Hossfeld D); Städtisches Krankenhaus, Kaiserslautern (Grosshaus A); Johanniter Krankenhaus Rheinhausen, Duisburg (Lang R); Kreiskrankenhaus Herford (Lange B); Städtisches Krankenhaus, Karlsruhe (Fischer J); Klinikum Stadt Ludwigshafen (Brass H); Krankenhaus Maria Hilf, Mönchengladbach (Reis E); St. Willehad Hospital, Wilhelmshaven (Augener W); University of Würzburg (Gieseler F). Address for reprints: W. Hiddemann, M.D., Ph.D., Department of Hematology and Oncology, GeorgAugust-University, Robert-Koch-Strasse 40, 37075 Göttingen, Germany. Received October 15, 1997; accepted January 12, 1998. 292 CANCER July 15, 1998 / Volume 83 / Number 2 FIGURE 1. The schedule for the sequential high-dose cytosine arabinoside and mitoxantrone protocol used in this study is shown. I nvasive fungal infections (IFIs) in patients receiving intensive myelosuppressive therapy for hematologic malignancies have increasingly become a matter of concern.1– 4 Especially in cases with prolonged neutropenia,5 systemic fungal infections may contribute substantially to infectious complications and early deaths. Measures for early detection and effective prophylactic strategies using active, nontoxic antifungal agents are therefore urgently needed. Until now, oral administration of amphotericin B suspension was the most frequently used approach. This treatment has been found to reduce the rate of IFI during chemotherapyinduced neutropenia6 but is hampered by moderateto-poor patient compliance and a lack of systemic activity. As a member of the new group of azole antifungals, fluconazole features a broad range of activity and has an excellent oral bioavailability.7 It causes few side effects and thereby offers a promising option for systemic antifungal prophylaxis in immunocompromised patients with severe, prolonged granulocytopenia. Therefore, the German AML Cooperative Group initiated a prospective randomized multicenter study to evaluate the impact of fluconazole prophylaxis on the incidence of proven or suspected IFI and the need for systemic antifungal therapy with amphotericin B in high risk patients with relapsed and refractory acute myeloid leukemia (AML). In preceding Phase II studies, patients undergoing intensive salvage therapy with sequential high dose cytosine arabinoside and mitoxantrone (S-HAM) for refractory or relapsed AML were found to experience a median duration of critical neutropenia of more than 5 weeks and an early death rate of 28%.8 These patients were therefore selected as a main target group for whom a beneficial effect of antifungal prophylaxis could be expected. METHODS Patients and Antileukemic Therapy Consecutive patients with relapsed and refractory AML who were admitted to the participating centers were eligible to participate in this study. Refractoriness against standard chemotherapy was defined according to previously established criteria,9 which included 1) primary resistance against 2 cycles of induction therapy, 2) first early relapse with a remission duration of less than 6 months, and 3) second and subsequent relapse. Patients with first relapses after 6 months of remission were not considered refractory to standard therapy and were included as relapsed AML. All patients were recruited from the first-line trials of the German AML Cooperative Group and had thus received standardized first-line treatment. For patients younger than 60 years, first-line therapy consisted of double induction therapy with either the repetitive application of the 9-day regimen of thioguanine, cytosine arabinoside (ara-C), and daunorubicin (TAD-9/TAD-9) or the sequential application of TAD-9 followed by high dose ara-C and mitoxantrone. Older patients all received one course of TAD-9 and were given a second TAD-9 course only when there was inadequate response to the first TAD-9 cycle. Patients of all ages who achieved a complete remission subsequently received TAD-9 for consolidation and monthly maintenance therapy for 3 years.10,11 Patients meeting the entry criteria were enrolled into the current study and were treated with S-HAM,12 which consisted of high dose ara-C 3 g/m2 or 1 g/m2 every 12 hours by a 3-hour infusion on Days 1, 2, 8, and 9 and mitoxantrone 10 mg/m2/day as a 30-minute infusion on Days 3, 4, 10, and 11, respectively (Fig. 1). The dose of ara-C was adjusted to the disease status according to the results of a previous trial that compared 3 g/m2 with 1 g/m2 ara-C in the treament of patients with refractory AML.8 Patients with refractory disease who were younger than 60 years received 3 g/m2 ara-C, whereas all other cases were treated with 1 g/m2 ara-C. To prevent photophobia and conjunctivitis induced by high dose ara-C, all patients received glucocorticoid eye drops every 6 hours starting before the first dose of the drug and continuing for 24 hours after the last dose. All patients also received granulocytecolony stimulating factor (G-CSF) 5mg/kg/day subcutaneously starting on Day 12 after the beginning of therapy. G-CSF was discontinued if a bone marrow examination on Day 18, i.e., 1 week after completion of S-HAM, revealed more than 5% residual leukemic blasts. In patients with adequate blast cell clearance Fluconazole Prophylaxis in AML/Kern et al. on Day 18, G-CSF was continued until the neutrophil count reached a value of more than 1500/mm3 for 3 consecutive days. Patients with antecedent hematologic disorders, secondary leukemias, and a preceding autologous or allogeneic bone marrow transplantation were excluded from the study. Other patients excluded were those with abnormal liver function tests (aspartate aminotransferase [AST], alanine aminotransferase [ALT], or alkaline phosphatase [AP] more than 3 times the upper normal limits, respectively; total bilirubin 2.0 mg/dL); impaired renal function (serum creatinine 2.0 mg/dL); severe infections; or pregnancy. Study Design and Antimicrobial Strategies On the basis of the above-mentioned antileukemic strategy, the current study aimed to assess the efficacy of fluconazole in addition to standard antimicrobial prophylaxis for the prevention of IFI and the need for empiric systemic antifungal therapy with amphotericin B. The study had the design of a prospective randomized multicenter analysis. Prior to the start of chemotherapy, patients were randomly assigned to either 1) systemic antifungal prophylaxis with fluconazole 400 mg administered orally (p.o.) daily in addition to standard antimicrobial prophylaxis consisting of co-trimoxazol 960 mg p.o. 3 times daily, colistine sulphate 2 million units p.o. 4 times daily, and amphotericin B suspension 40 mg p.o. 6 times daily; or 2) standard antimicrobial prophylaxis only. To avoid imbalances in the risk profile, patients were stratified according to age (, or $ 60 years) and disease status (primary refractoriness, relapse after #6 months of first remission, relapse after 6 months and #18 months of first remission, relapse after .18 months of first remission, or second/subsequent relapse).13 Antimicrobial prophylaxis was continued until parenteral antimicrobial therapy was required or until a leukocyte count of more than 1000/mL was reached. Parenteral antimicrobial therapy was initiated on the occurence of fever of unknown origin (FUO) or fever with clinical and/or microbiologically verified infection. Antimicrobial treatment for FUO comprised two drug combinations of an aminoglycoside with either a third-generation cephalosporin or ureidopenicillin. On persistence or recurrence of fever, a combination of carbapenems with glycopeptides was initiated. In addition, intravenous antifungal therapy was begun. This comprised amphotericin B in patients who were randomized to fluconazole prophylaxis. In patients randomized to the control, intravenous fluconazole was given instead, which was replaced by amphotericin B only when there was no response. Patients with pneumonia, however, were uniformly started on am- 293 photericin B in addition to antibiotic therapy. The addition of other antifungal agents was dependent on the patient’s clinical condition and was left at the discretion of the responsible physician. Study Parameters Infectious complications were classified according to the Consensus Report of the Immunocompromised Host Society,14 as 1) FUO not accompanied by either clinical or microbiologic evidence of infection; 2) clinically defined infections referring to the diagnosis of a site of infection without determination of the infectious agent; or 3) microbiologically defined infections consisting of bacteremia, fungemia, a microbiologically defined site of infection, or a combination of the three. In addition, IFIs were classified according to criteria suggested by Behre et al. (Table 1).15 Toxicity was evaluated according to the World Health Organization (WHO) grading system.16 Response to therapy was assessed according to Cancer and Leukemia Group B criteria.17 Complete remission (CR) was defined as normal cellular bone marrow with normal erythroid and myeloid elements as well as myeloblasts, promyelocytes, and other leukemic cells totaling less than 5%, and with normal peripheral blood platelet and white blood cell counts for at least 4 weeks. Patients who had more than 5% myeloblasts but fewer than 25% blasts, with otherwise normal bone marrow, were considered to be in partial remission (PR), as were patients who fulfilled the criteria of CR except for full recovery of peripheral blood platelet and/or white blood cell counts. Patients with persistent leukemic blasts in the bone marrow or blood or with leukemic regrowth within 4 weeks after initial response were considered nonresponders (NRs). Patients dying within 6 weeks after completion of antileukemic therapy without evidence of leukemic regrowth were classified as early deaths (EDs). The duration of critical cytopenia was evaluated by the time of leukocyte recovery to more than 1000/mL from the onset of S-HAM treatment. The time to CR was measured from the onset of treatment to the date of documented CR, and disease free survival was measured from the date of documented CR to relapse or death during remission. Survival was measured as the time from the beginning of treatment to death. Time to treatment failure was measured as the time from the beginning of treatment to death without evidence of leukemia, documentation of persistent leukemia, or relapse. Statistics The primary end point of the current study was the impact of fluconazole prophylaxis in addition to stan- 294 CANCER July 15, 1998 / Volume 83 / Number 2 TABLE 1 Definition of Invasive Fungal Infections in Neutropenic Patients Invasive pulmonary aspergillosis Proven Histology 6 culture Probable Pneumonia or other organ infection unresponsive to antibiotics, with no other microbiologically documented causative organism and one of the following conditions: Culture of normally sterile tissues, bronchoalveolar lavage, blood sputum, or nose Repeated antigen detection or increasing antibody titers in blood Pulmonary lesions with halo sign or air crescent sign in computed tomography scan Possible Fever unresponsive to antibiotics, with no other microbiologically documented causative organism and one of the following conditions: Culture of normally sterile tissues, bronchoalveolar lavage, blood sputum, or nose Repeated antigen detection or increasing antibody titers in blood Invasive candidiasis Proven Histology 6 culture or Histology or culture of normally sterile tissues, including blood, with fever unresponsive to antibiotics and no other microbiologically documented causative organism Probable Pneumonia or other organ infection unresponsive to antibiotics, with no other microbiologically documented causative organism and one of the following conditions: Culture or bronchoalveolar lavage Increasing antigen or antibody titers in blood Multiple hepatic and splenic lesions in computed tomography scan/sonogram Possible Fever unresponsive to antibiotics with no other microbiologically documented causative organism and one of the following conditions: Culture, excluding blood Increasing antigen or antibody titers in blood Possible invasive fungal infection No invasive pulmonary aspergillosis or candidiasis as defined above, but one of the following conditions: Pneumonia unresponsive to antibiotics and/or responsive to antimycotics 6 antibiotics Fever unresponsive to antibiotics and responsive to antimycotics 6 antibiotics dard antimicrobial prophylaxis on the incidence of IFI and the requirement of additional systemic antifungal therapy with amphotericin B as compared with a randomly assigned control group receiving standard antimicrobial prophylaxis alone. The secondary end point was the ED rate. Assuming a reduction of 20% in the incidence of IFI or the requirement of systemic antifungal therapy by fluconazole prophylaxis, with a 5 0.05 and b 5 0.20, it was anticipated that 73 patients would be enrolled into each treatment arm. Numeric values were compared by the chi-square test, Fisher’s exact test, and Student’s t test. Remission duration and survival were calculated for the fluconazole and control groups according to Kaplan–Meier estimates, respectively, and were compared using the log rank test. After enrollment of 75 patients into the study, 68 of whom could be fully evaluated, a scheduled interim analysis disclosed a difference between the study and control groups in the number of patients who required systemic amphotericin B. Based on these data, a hypothetical analysis of the total of 146 originally planned patients was performed, assuming that none of the following patients with fluconazole prophylaxis and 50% of the following patients without fluconazole prophylaxis would receive amphotericin B. Under this best-case assumption, no significant advantage for fluconazole prophylaxis could be detected. Therefore, it was decided to stop further recruitment of patients into the trial. Study Conduct Prior to receiving therapy, all patients gave their informed consent for participation in the current evaluation after having been advised about the purpose and investigational nature of the study as well as the potential risks of participation. The study design adhered to the Declaration of Helsinki and was approved by the ethics committees of the participating institutions prior to its initiation. RESULTS Patient Characteristics Between May 1992 and January 1996, 84 patients entered the study, 68 of whom could be fully evaluated. Causes for exclusion from analysis were secondary leukemia (for 3 patients), application of the study drug before randomization (for 2 patients), treatment not in accordance with the results of the randomization (for 1 patient), discontinuation of the S-HAM chemotherapy after the first 4 days (for 2 patients), and incomplete documentation (for 8 patients). Of the patients who could be evaluated, 36 and 32 were randomized to the fluconazole arm and the control arm, respectively. The patient characteristics and risk profiles of the two groups were comparable. The patients’ ages ranged from 21 to 73 and from 17 to 72 years (median, 52 and 45 years) for the fluconazole prophylaxis and control groups (P . 0.05), respectively. All patients had received prior chemotherapy for their disease as indicated above. Two patients (6%) in each group were younger than 60 years and had AML refractory to first-line therapy. Early and late relapses after a first complete remission of # and 6 months’ duration occurred in 7 (19%) and 19 (53%) vs. 7 (22%) and 18 (56%) patients younger than 60 years, respectively. The patients were older than 60 years in 8 (22%) vs. 4 Fluconazole Prophylaxis in AML/Kern et al. TABLE 2 Patient Characteristics and Risk Factors Age, yrs (median/range) Gender (male/female) Disease status age ,60 yrs 1 refractory AML age ,60 yrs 1 CR duration #6 mos age ,60 yrs 1 CR duration .6 mos age ,60 yrs 1 2nd relapse age .60 yrs Antifungal treatment during preceeding neutropenias for invasive pulmonary aspergillosis for invasive candidiasis for suspected IFI for FUO Infections at study entry FUO Gastrointestinal tract infection Soft tissue infection Recovery of leukocytes to .1000/mL (days after start of chemotherapy; median/range) Time to complete remission (days after start of chemotherapy; median/range) Duration of fluconazole prophylaxis (days; median/range) 295 Toxicity of Antileukemic Therapy Fluconazole (n 5 36) Control (n 5 32) 52/21–73 20/16 45/17–72 17/15 2 (6%) 7 (19%) 19 (53%) — 8 (22%) 2 (6%) 7 (22%) 18 (56%) 1 (3%) 4 (13%) 30 (83%) 2 3 15 10 7 (19%) 6 1 — 24 (75%) 1 3 14 6 2 (6%) 1 — 1 35/22–58 34/19–59 51/30–71 57/37–93 23/2–56 — AML: acute myeloid leukemia; CR: complete remission; IFI: invasive fungal infection; FUO: fever of unknown origin. (13%) cases (P . 0.05) (Table 2). All patients who could be evaluated received one course of S-HAM therapy. During preceding neutropenias, systemic antifungal therapy was given to 30 patients (83%) versus 24 patients (75%), in 9 of whom IFI had been documented (Aspergillus species, 2 vs. 1; Candida species, 3 vs. 3). Antifungal therapy had been initiated for suspected IFI and for FUO resistant to antibiotic treatment in 15 and 10 versus 14 and 6 cases, respectively. Infections at study entry were more frequent within the fluconazole prophylaxis group (7 patients [19%] versus 2 patients [6%]), mainly due to a higher prevalence of FUO (6 [17%] vs. 1 [3%]; P 5 0.07). The median duration of critical cytopenia was similar in both groups, with the recovery of leukocytes to more than 1000/mL occurring at a median of 35 versus 34 days after the start of therapy (Table 2). Consequently, there was no significant difference in the median time to CR (51 vs. 57 days after the start of therapy; P 5 0.06; Fig. 2). Fluconazole prophylaxis was carried out for a median of 23 days (range, 2–56 days). Severe nonhematologic toxicities (WHO Grade 3/4) during S-HAM therapy were not dependent on the use of fluconazole prophylaxis and consisted mainly of nausea/vomiting, stomatitis, and diarrhea (Table 3). Furthermore, mild or severe elevation of liver enzymes and parameters for cholestasis were not different between the fluconazole prophylaxis group and the control group. Infectious Episodes The median number of febrile episodes was 1 in both groups (range, 0 –3), whereas the total number of febrile days was higher in the group receiving fluconazole prophylaxis (median, 9 vs. 6 days; range, 0 –29 vs. 0 – 60 days; P . 0.05; Table 4). The median time to the first febrile episode was 10 versus 15 days after the start of therapy (P . 0.05, Fig. 3). The infectious complications that were encountered during S-HAM therapy are summarized in Table 4. Three patients remained free of infections throughout the whole study period. No major differences between the two study groups were observed for the incidence of FUO (22 episodes [61%] vs. 16 episodes [50%]) or clinically defined infections (20 [56%] vs. 16 [50%]). There was a trend toward more infections of the gastrointestinal tract among patients who received fluconazole prophylaxis; this trend was not statistically significant (9 vs. 3; P 5 0.08). Marked differences were seen in the incidence of microbiologically defined infections (18 [50%] vs. 10 [31%]; P 5 0.09), mainly due to a strong trend toward more bacteremias in the fluconazole group (15 [42%] vs. 7 [22%]; P 5 0.07), which reached significance for gram positive bacteremias (33% vs. 13%; P 5 0.04). Bacteremias also occurred earlier in this group. The frequencies of all other categories of infection were similar for both groups. Overall, 12 patients (33%) versus 9 patients (28%) developed pneumonia, which occurred earlier in patients who received fluconazole prophylaxis. In total, causative organisms were predominantly gram positive and gram negative bacteria identified in 26 and 14 cases, respectively. Fungal infections were documented in only 4 patients (Table 5). There were 2 versus 2 cases with proven invasive candidiasis, whereas invasive pulmonary aspergillosis was not observed. Possible IFI occurred in 19 (53%) versus 11 (34%) patients (P 5 0.10; Table 6). Antimicrobial Therapy Antifungal therapy was initiated in 22 (61%) versus 18 (56%) patients (Table 7). Systemic amphotericin B was 296 CANCER July 15, 1998 / Volume 83 / Number 2 FIGURE 2. Time to complete remission is shown. TABLE 3 Nonhematologic Toxicity TABLE 4 Infectious Complications Fluconazole (n 5 36) Nausea/vomiting Stomatits Diarrhea Bilirubin AP AST/ALT Bleeding Creatinine Lung Allergy Cardiac rhythm Cardiac function Pericarditis Peripheral nervous system Consciousness Control (n 5 32) WHO 3/4 WHO 3/4 WHO 3/4 WHO 3/4 5 (14%) 13 (36%) 8 (22%) 10 (28%) 7 (19%) 3 (8%) 11 (31%) 7 (19%) — 3 (8%) 3 (8%) 1 (3%) — 14 (39%) 5 (14%) 6 (17%) 1 (3%) — 2 (6%) — — 1 (3%) — 1 (3%) 1 (3%) — 10 (31%) 10 (31%) 5 (16%) 4 (13%) 2 (6%) 5 (16%) 2 (6%) 1 (3%) — — 2 (6%) — 1 (3%) 7 (22%) 5 (16%) 5 (16%) 3 (9%) — 1 (3%) 2 (6%) — 2 (6%) — 2 (6%) 1 (3%) — 1 (3%) 1 (3%) — 1 (3%) — — — 1 (3%) WHO: World Health Organization; AP: alkaline phosphatase; AST: aspartate aminotransferase; ALT: alanine aminotransferase. given to more patients in the fluconazole prophylaxis group (20 [56%] vs. 9 [28%]). Seven patients within the control group received fluconazole intravenously as initial antifungal therapy, none of which required subsequent parenteral amphotericin B. The median number of antibiotic regimens was 3 versus 2 (range, 0 –7 vs. 0 – 6). The antimicrobial therapy was successful in 25 (71%) versus 21 (70%) of 35 versus 30 patients with infections (Table 8), whereas in 3 patients (9%) versus 3 patients (10%) only an incomplete control of the infections was obtained. Fatal infectious complications occurred in 7 patients (19%) versus 6 patients No. of febrile episodes (median/range) No. of febrile days (median/range) Time to first infection (days; median/range) No infection FUO Clinically defined infections Pneumonia Sepsis syndrome Septic shock Gastrointestinal tract infection Perianal infection Catheter-related infection Other Microbiologically defined infections Bacteriemia Gram positive Gram negative Fungemia Pneumonia Sepsis syndrome Septic shock Gastrointestinal tract infection Perianal infection Catheter related infection Other Fluconazole (n 5 36) Control (n 5 32) 1/0–3 9/0–29 10/0–33 1(3%) 22 (61%) 20 (56%) 6 (17%) — 2 (6%) 9 (25%) — 1 (3%) 2 (6%) 18 (50%) 15 (42%) 12 (33%) 3 (8%) 2 (6%) 6 (17%) 2 (6%) 2 (6%) 1 (3%) — 1 (3%) 2 (6%) 1/0–3 6/0–60 15/0–22 2 (6%) 16 (50%) 16 (50%) 6 (19%) — 1 (3%) 3 (9%) 3 (9%) 2 (6%) 1 (3%) 10 (31%) 7 (22%) 4 (13%) 3 (9%) 1 (3%) 3 (9%) 2 (6%) 2 (6%) 2 (6%) — 2 (6%) 1 (3%) FUO: fever of unknown origin. (19%), in 4 of whom fungi or bacteria were documented to be the causative organism. Overall Response and Antileukemic Activity of S-HAM Corresponding to the similarity of results on the frequency and severity of infectious complications, no significant differences were observed in overall re- Fluconazole Prophylaxis in AML/Kern et al. 297 FIGURE 3. Time to the first febrile episode is shown. TABLE 5 Isolated Causative Organisms Gram positive bacteria Gram negative bacteria Fungi Fluconazole (n 5 26) Control (n 5 19) Total (n 5 44) 17 (68%) 6 (24%) 2 (8%) 9 (47%) 8 (42%) 2 (11%) 26 (59%) 14 (32%) 4 (9%) sponse, disease free survival, and overall survival. The rates of ED were 22% versus 19%, respectively; 61% versus 50% of patients achieved a CR and 17% versus 31% had NR (Table 9). Median time to treatment failure was 5.2 versus 3.2 months, median disease free survival for patients in complete remission was 6.7 versus 5.1 months, and median total survival was 7.3 versus 6.8 months (Fig. 4). DISCUSSION The German AML Cooperative Group conducted the current study with the aim of reducing the incidence of IFI and the requirement of systemic antifungal therapy with amphotericin B by administering antifungal prophylaxis with fluconazole in addition to standard antimicrobial therapy to patients receiving intensive chemotherapy for high risk AML. In these patients, infectious complications were often encountered, a considerable proportion of which were proven or suspected to be of fungal origin. Furthermore, the high ED rate of more than 25% demands improved antimicrobial therapy. Among the candidate drugs, fluconazole was the most attractive because of its pharmacokinetic profile, its broad range of antifungal activity, and its good tolerability.18 –21 Fluconazole prophylaxis had already been as- sessed in randomized studies of patients receiving intensive chemotherapy or bone marrow transplantation mainly for acute leukemias. The majority of these trials revealed that fluconazole significantly reduced the incidence of candida colonization but was not superior in the prevention of IFI.22–30 Only a few placebo-controlled studies have evaluated the prophylactic antifungal efficacy of fluconazole. Two early trials involved heterogenous collectives of oncologic patients receiving chemotherapy, radiotherapy, antibiotic therapy, or glucocorticoid therapy.31,32 Fluconazole significantly reduced the incidence of oropharyngeal candidiasis among those patients but had no effect on the occurrence of IFI. This negative result was thought to emerge from the application of low doses of 50 mg fluconazole only, which was well below currently used doses. In fact, in two subsequent trials, the daily administration of 400 mg of fluconazole not only reduced the incidence of not only superficial fungal infections but also IFIs in patients undergoing bone marrow transplantation. In one of these studies, reduced overall mortality until Day 110 after transplantation was observed in the fluconazole arm.33,34 Winston et al. were the first to provide data on prophylaxis with fluconazole (400 mg daily) in patients receiving intensive chemotherapy mainly for acute leukemias.35 In a randomized multicenter study, a significant decrease in fungal infections was shown in the fluconazole arm (9% vs. 21%; P 5 0.02), mainly due to a reduction in superficial fungal infections. At the end of the study period, the colonization of patients with Candida albicans was also reduced significantly. However, neither the incidence of IFI (4% vs. 8%; P 5 0.30) nor the frequency of the application of systemic amphotericin B (64% vs. 74%; P 5 0.10) were 298 CANCER July 15, 1998 / Volume 83 / Number 2 TABLE 6 Fungal Infections Fluconazole Invasive fungal infection Invasive candidiasis Invasive aspergillosis Control Proven Probable Possible Proven Probable Possible 2 — — — 19 — — 2 — — — 11 — — TABLE 7 Antimicrobial Therapy TABLE 9 Antileukemic Activity of S-HAM P value Systemic antifungal therapy Amphotericin B Fluconazole 5-Flucytosine Itraconazole Ambisome No. of antibiotic regimens (median/range) 0.02 — 0.07 0.07 0.72 Fluconazole (n 5 36) Control (n 5 32) 22 (61%) 20 (56%) — 15 (42%) 1 (3%) 1 (3%) 3/0–7 18 (56%) 9 (28%) 7 (22%) 7 (22%) 5 (16%) 1 (3%) 2/0–6 Complete remission Nonresponse Early death Time to treatment failure (mos, median) Disease free survival (mos, median) Total survival (mos, median) Fluconazole (n 5 36) Control (n 5 32) Total (n 5 68) 22 (61%) 6 (17%) 8 (22%) 5.2 6.7 7.3 16 (50%) 10 (31%) 6 (19%) 3.2 5.1 6.8 38 (56%) 16 (24%) 14 (21%) 4.8 5.5 7.2 S-HAM: sequential high dose cytosine arabinoside and mitoxantrone. TABLE 8 Efficacy of Antimicrobial Therapy Result of therapy Fluconazole (n 5 35) Control (n 5 30) Complete remission Improvement Death due to infection Septic shock Candida krusei Candida tropicalis Streptococcus viridans Enterococcus species Pneumonia 25 (71%) 3 (9%) 7 (20%) 5 1 — — 1 2 20 (70%) 3 (10%) 6 (20%) 3 — 1 1 — 3 influenced by the fluconazole prophylaxis. As a result, the mortality was the same for both groups (21% vs. 18%). Another analysis was performed by Schaffner et al. in a randomized single-center study evaluating antifungal prophylaxis with fluconazole (400 mg daily) in patients receiving chemotherapy for acute leukemias and high grade lymphomas.36 These authors also found a significant reduction in the incidence of oropharyngeal candidiasis (1% vs. 12%; P 5 0.01) as well as a reduction in colonization with Candida species (8% vs. 37%; P , 0.0001) in patients who received fluconazole prophylaxis. In contrast to the results obtained by Winston et al., a trend toward a reduced need for systemic amphotericin B was reported (33% vs. 47%; P 5 0.08), which reached statistical significance for patients with FUO (16% vs. 30%; P 5 0.04). However, as in the above-mentioned study, neither the incidence of IFI (11% vs. 11%) nor the mortality rate (5% vs. 7%) could be diminished by fluconazole prophylaxis. Furthermore, the significance of this analysis was restricted by the relatively short duration of neutropenia (median, 22 days), thereby raising the question of whether immunosuppression was severe enough to allow fluconazole prophylaxis to be effective. Further evidence of the efficacy of fluconazole prophylaxis came from a placebo-controlled randomized single-center study37 of a series of patients undergoing allogeneic or autologous bone marrow transplantation or intensive chemotherapy for acute leukemia or blast crisis of chronic myelogenous leukemia. This study revealed a significant reduction in the need for systemic amphotericin B therapy (22% vs. 58%; P , 0.01). In addition, the number of days with fever was significantly diminished (5 days vs. 9 days; P , 0.05). Besides by the heterogeneity of the study population, the results of this study were compromised by a relatively short duration of critical neutropenia (only 16 days). As could have been anticipated, no differences were observed in either the incidence of IFI or the mortality rate. Taking the shortcomings of the above-mentioned analyses into account, the current study was carried out with a homogenous group of patients with advanced AML who all received an identical type of intensive antileukemic therapy and standardized an- Fluconazole Prophylaxis in AML/Kern et al. 299 FIGURE 4. Total survival is shown. timicrobial prophylaxis and intervention. These patients were chosen for the current approach because they faced a prolonged period of severe neutropenia regularly exceeding 30 –35 days and a high risk of lethal complications from severe infections in the range of 25–30%. The randomization resulted in a similarity in the patients’ risk profiles between the two groups, excluding potential biases by differences in disease status, prior infections, infections at study entry, or duration of critical neutropenia. All patients randomized to receive fluconazole tolerated the drug very well for a median duration of 23 days. Whereas the time to the occurrence of infectious episodes and their duration as well as the incidences of FUO and of clinically defined infections did not differ between the two groups, an excess of bacteremias led to a higher frequency of microbiologically defined infections in patients receiving fluconazole prophylaxis. Systemic antifungal therapy with amphotericin B was initiated more often in the group with fluconazole prophylaxis, whereas fluconazole was used successfully as interventional therapy in seven patients of the control group. There were seven and six cases with fatal infections in the two groups, respectively. No differences in the antileukemic activity of the S-HAM regimen were observed between the two groups. Overall, the infectious complications manifested themselves similarly in both study arms. In particular, pneumonia which is a major threat to patients with prolonged severe immunosuppression38 was not less frequent among patients with fluconazole prophylaxis (33% vs. 28%). More episodes of bacteremia were encountered in the fluconazole prophylaxis group (42% vs. 22%; P 5 0.07), mainly due to gram positive bacteria (33% vs. 13%; P 5 0.04). Similar observations had been made in two previous studies.20,36 Schaffner et al. also detected more bacteremias during fluconazole prophylaxis (36% vs. 21%; P 5 0.04), the difference being mainly due to gram negative bacteria (15% vs. 5%; P 5 0.05). In a placebo-controlled trial, Palmblad et al. observed more bacteremias during ketoconazole prophylaxis (74% vs. 37%; P 5 0.0001). However, it was unclear whether there was a causal connection between the prophylaxis and the bacteremia or whether there were differences between the study and control groups with regard to patients’ conditions. The current study failed to detect a difference in the incidence of IFI between the study group and the control group, and these results were in accordance with previous analyses.35–37 This was based in part on the overall low rate of proven IFI and reflected the difficulties of this diagnosis when the rigid criteria suggested by Behre et al were used.15 A major and yet not previously reported finding of the current study was an excess use of systemic amphotericin B therapy in patients receiving fluconazole prophylaxis. Because the analysis of the study was one-sided, only a possible advantage of fluconazole could be analyzed, and no determination of the significance of the observed difference favoring the control group could be performed. However, this finding probably resulted from the general strategy of antifungal therapy that was used in this study. In most other trials, patients with fever or infections and no response to antibiotics for 3– 6 days usually received amphotericin B as antifungal therapy.33–37 In contrast, patients within the control group of the current study were scheduled to receive intravenous fluconazole for systemic antifungal treatment when FUO persisted for 3 days during antibiotic therapy. Systemic amphotericin B was initiated only if FUO persisted for 3 more 300 CANCER July 15, 1998 / Volume 83 / Number 2 days. As a result, seven patients within the control group received antifungal therapy with fluconazole, none of whom had to proceed to systemic amphotericin B. This finding suggests that, at least in FUO, fluconazole may be effective and thus replace the more toxic treatment with systemic amphotericin B. Furthermore, these results suggest that the need for systemic amphotericin B therapy may have been overestimated in previous investigations. In fact, a more recent trial by the Paul Ehrlich Society for Chemotherapy emphasized that fluconazole serves as a valuable therapeutic alternative to intravenous amphotericin B for interventional therapy in patients with FUO that is resistant to antibiotic regimens.39 This finding prompts reconsideration of the most appropriate circumstances for fluconazole treatment and favors its application as an intervention rather than an antifungal prophylaxis. This conclusion is supported by the potential risks of fluconazole prophylaxis, including breakthrough candidemias and an increasing incidence of infections due to Candida non-albicans species.40 – 44 However, additional trials addressing the use of fluconazole as preemptive antifungal therapy in neutropenic patients with FUO persisting after 3 days of antibiotic therapy must be conducted to determinate the appropriate role of this drug in this setting. Given the secondary end point of the current trial, with fluconazole failing to reduce the mortality rate within the study group, the general use of the drug as antifungal prophylaxis cannot be supported. Rather, fluconazole as a potent antifungal agent should be an option for interventional therapy for patients with suspected or proven IFI. This strategy seems appropriate at least for nonneutropenic patients, for whom fluconazole was shown to be aseffective as intravenous amphotericin B in the treatment of candidemia in a randomized study.45 Further evidence of the efficacy of fluconazole as interventional antifungal therapy for neutropenic patients came from a matched cohort study in which cancer patients with candidemia treated with fluconazole had the same response rate as those who received amphotericin B therapy.46 The validity of these results for neutropenic patients in general is hampered by a proportion of only 24% of the patients’ being neutropenic at the time of enrollment in the study. Therefore, further assessment of the applicability of fluconazole to severely immunocompromised neutropenic patients with suspected or proven IFI should focus on randomized comparisons with intravenous amphotericin B, which has hitherto comprised the standard interventional antifungal therapy. REFERENCES 1. Bow EJ, Loewen R, Cheang MS, Schacter B. Invasive fungal disease in adults undergoing remission-induction therapy for acute myeloid leukemia: the pathogenetic role of the antileukemic regimen. Clin Infect Dis 1995;21:361–9. 2. DeGregorio MW, Lee WMF, Linker CA, Jacobs RA, Ries CA. Fungal infections in patients with acute leukemia. Am J Med 1982;73:543– 8. 3. Donhuisen K, Samandari S. Tiefe mykosen in patienten mit leukämien und lymphomen. Dtsch Med Wochenschr 1985; 110:903–7. 4. Uzun O, Anaissie EJ. Antifungal prophylaxis in patients with hematologic malignancies: a reappraisal. Blood 1995; 86:2063–72. 5. Link H, Maschmeyer G, Meyer P, Hiddemann W, Stille W, Helmerking M, et al. Interventional antimicrobial therapy in febrile neutropenic patients. Study Group of the Paul Ehrlich Society for Chemotherapy. Ann Hematol 1994;69:231– 43. 6. Ezdinli EZ, O’Sullivan DD, Wasser LP, Kim U, Stutzmann L. Oral amphotericin for candidiasis in patients with hematologic neoplasms. JAMA 1979;242:258 – 60. 7. Goa KL, Barradell LB. Fluconazole: an update of its pharmacodynamic and pharmacokinetic properties and therapeutic use in major superficial and systemic mycoses in immunocompromised patients. Drugs 1995;50:658 –90. 8. Kern W, Aul C, Maschmeyer G, Schönrock-Nabulsi R, Ludwig WD, Bartholomäus A, et al. Superiority of high-dose over intermediate-dose cytosine arabinoside (AraC) in the treatment of patients with high-risk acute myeloid leukemia (AML). Ann Hematol 1997;74(2 Suppl 1):171. 9. Hiddemann W, Martin WR, Sauerland CM, Heinecke A, Büchner T. Definition of refractoriness against conventional chemotherapy in acute myeloid leukemia: a proposal based on the results of retreatment of thioguanine, cytosine arabinoside, and daunorubicin (TAD 9) in 150 patients with relapse after standardized first line therapy. Leukemia 1990;4: 184 – 8. 10. Büchner T, Urbanitz D, Hiddemann W, Ruhl H, Ludwig WD, Fischer J, et al. Intensified induction and consolidation with or without maintenence chemotherapy for acute myeloid leukemia (AML): two multicenter studies of the German AML Cooperative Group. J Clin Oncol 1985;3:1583–9. 11. Büchner T, Hiddemann W, Schaefer UW, Löffler H, Maschmeyer G, Ludwig WD, et al. Combined effect of very early intensification and prolonged post-remission chemotherapy in patients with AML. Leukemia 1992;6 (Suppl 4):68 –70. 12. Kern W, Schleyer E, Unterhalt M, Wörmann B, Büchner T, Hiddemann W. High antileukemic activity of sequential high-dose cytosine arabinoside and mitoxantrone (S-HAM) in refractory acute leukemias: results of a clinical phase II study. Cancer 1997;79:59 – 68. 13. Hiddemann W, Martin WR, Sauerland CM, Heinecke A, Büchner T. Definition of refractoriness against conventional chemotherapy in acute myeloid leukemia: a proposal based on the results of retreatment of thioguanine, cytosine arabinoside, and daunorubicin (TAD 9) in 150 patients with relapse after standardized first line therapy. Leukemia 1990;4: 184 – 8. 14. The Immunocompromised Host Society. The design, analysis, and reporting of clinical trials on the empirical antibiotic management of the neutropenic patient: report of a consensus panel. J Infect Dis 1990;161:397– 401. Fluconazole Prophylaxis in AML/Kern et al. 15. Behre GF, Schwartz S, Lenz K, Ludwig WD, Wandt H, Schilling E, et al. Aerosol amphotericin B inhalations for prevention of invasive pulmonary aspergillosis in neutropenic cancer patients. Ann Hematol 1995;71:287–91. 16. World Health Organization. A handbook for reporting results of cancer treatment. WHO Publication No. 38. Geneva: World Health Organization, 1979. 17. Yates J, Glidewell O, Wiernik P, Cooper MR, Steinberg D, Dosik H, et al. Cytosine arabinoside with daunorubicin or adriamycin for therapy of acute myelocytic leukemia: a CALGB study. Blood 1982;60:454 – 62. 18. Donelly JP, Starke ID, Galton DAG, Catovsky D, Goldman JM, Darrell JH. Oral ketoconazole and amphotericin B for the prevention of yeast colonisation in patients with acute leukemia. J Hosp Infect 1984;5:83–91. 19. Hann IM, Corringham R, Keaney M, Blacklock HA, Keaney M, Shannon M, et al. Ketoconazole versus nystatin plus amphotericin B for fungal prophylaxis in severely immunocompromised patients. Lancet 1982;333:826 –9. 20. Palmblad J, Lönnqvist B, Carlsson B, Grimfors G, Jarnmark M, Lerner R, et al. Oral ketokonazole prophylaxis for candida infections during induction therapy for acute leukemia in adults: more bacteraemias. J Intern Med 1992;231:363–70. 21. Perfect JR, Klotman ME, Gilbert CC, Crawford DD, Rosner GL, Wright KA, et al. Prophylactic intravenous amphotericin B in neutropenic autologous bone marrow transplant recipients. J Infect Dis 1992;165:891–7. 22. Annaloro C, Oriana A, Tagliaferri E, Bertolli V, Della-Volpe A, Soligo D, et al. Efficacy of different prophylactic antifungal regimens in bone marrow transplantation. Haematologica 1995;80:512–7. 23. Bodey GP, Anaissie EJ, Elting LS, Estey E, O’Brien S, Kantarjian H. Antifungal prophylaxis during remission induction therapy for acute leukemia fluconazole versus intravenous amphotericin B. Cancer 1994;73:2099 –106. 24. Egger T, Gratwohl A, Tichelli A, Uhr M, Stebler-Gysi C, Passweg J, et al. Comparison of fluconazole with oral polyenes in the prevention of fungal infections in neutropenic patients: a prospective, randomized, single-center study. Support Care Cancer 1995;3:139 – 46. 25. Ellis ME, Clink H, Ernst P, Halim MA, Padmos A, Spence D, et al. Controlled study of fluconazole in the prevention of fungal infections in neutropenic patients with haematologic malignancies and bone marrow transplant recipients. Eur J Clin Microbiol Infect Dis 1994;13:3–11. 26. Menichetti F, Del Favero A, Martino P, Bucaneve G, Micozzi A, D’Antonio D, et al. Preventing fungal infection in neutropenic patients with acute leukemia: fluconazole compared with oral amphotericin B. The GIMEMA Infection Program. Ann Intern Med 1994;120:913– 8. 27. Meunier F, Aoun M, Janssens M, Dekoster C, Paesmans M. Chemoprophylaxis of fungal infections in granulocytopenic patients using fluconazole vs. oral amphotericin B. Drug Invest 1991;3:258 – 65. 28. Ninane J and a multicenter study group. A multicenter study of fluconazole versus oral polyenes in the prevention of fungal infections in children with hematological or oncological malignancies. Eur J Clin Microbiol Infect Dis 1994;13:6 –13. 29. Philpott-Howard JN, Wade JJ, Mufti GJ, Brammer KW, Ehninger G. Randomized comparison of oral fluconazole versus oral polyenes for the prevention of fungal infection in patients at risk of neutropenia. J Antimicrob Chemother 1993;31:973– 84. 30. Rozenberg-Arska M, Dekker AW, Branger J, Verhoef J. A ran- 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 301 domized study to compare oral fluconazole to amphotericin B in the prevention of fungal infections in patients with acute leukemia. J Antimicrob Chemother 1991;27:369–76. Bodey GP, Samonis G, Rolston K. Prophylaxis of candidiasis in cancer patients. Semin Oncol 1990;17 (Suppl 6):24 – 8. Samonis G, Rolston K, Karl C, Miller P, Bodey GP. Prophylaxis of oropharyngeal candidiasis with fluconazole. Rev Infect Dis 1990;12:369 –73. Slavin MA, Osborne B, Adams R, Levenstein MJ, Schoch HG, Feldman AR, et al. Efficacy and safety of fluconazole prophylaxis for fungal infections after marrow transplantation: a prospective, randomized, double-blind study. J Infect Dis 1995;171:1545–52. Goodman JL, Winston DJ, Greenfield RA, Chandrasekar PH, Fox B, Kaizer H, et al. A controlled trial of fluconazole to prevent fungal infections in patients undergoing bone marrow transplantation. N Engl J Med 1992;326:845–51. Winston DJ, Chandrasekar PH, Lazarus HM, Goodman JL, Silber JL, Horowitz H, et al. Fluconazole prophylaxis of fungal infections in patients with acute leukemia. Ann Intern Med 1993;118:495–503. Schaffner A, Schaffner M. Effect of prophylactic fluconazole on the frequency of fungal infections, amphotericin B use, and health care costs in patients undergoing intensive chemotherapy for hematologic neoplasias. J Infect Dis 1995;172:1035– 41. Chandrasekar PH, Gatney CM, Bone Marrow Transplantation Team. Effect of fluconazole prophylaxis on fever and use of amphotericin in neutropenic cancer patients. Chemotherapy 1994;40:136 – 43. Maschmeyer G, Link H, Hiddemann W, Meyer P, Helmerking M, Eisenmann E, et al. Pulmonary infiltrations in febrile patients with neutropenia: risk factors and outcome under empirical antimicrobial therapy in a randomized multicenter study. Cancer 1994;73:2296 –304. Link H, Hiddemann W, Maschmeyer G, Buchheidt D, Glass B, Cornely O, et al. Antimicrobial therapy in neutropenic patients with unexplained fever, PEG Study II. Abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Canada. Washington, DC: ASM Press, 1997:372. Johnson EM, Warnock DW. Azole drug resistance in yeasts. J Antimicrob Chemother 1995;36:751–5. Nguyen MH, Peacock JE, Morris AJ, Tanner DC, Nguyen ML, Snydman DR, et al. The changing face of candidemia: emergence of non-candida albicans species and antifungal resistance. Am J Med 1996;100:617–23. Price MF, LaRocco MT, Gentry LO. Fluconazole susceptibilities of candida species and distribution of species recovered from blood cultures over a 5-year period. Antimicrob Agents Chemother 1994;38:1422– 4. Rex JH, Rinaldi MG, Pfaller MA. Resistance of candida species to fluconazole. Antimicrob Agents Chemother 1995;39:1– 8. Wingard JR, Merz WG, Rinaldi MG, Miller CB, Karp JE, Saral R. Association of torulopsis glabrata infections with fluconazole prophylaxis in neutropenic bone marrow transplant patients. Antimicrob Agents Chemother 1993;37:1847–9. Rex JH, Bennett JE, Sugar AM, Pappas PG, van der Horst CM, Edwards JE, et al. A randomized trial comparing fluconazole with amphotericin B for the treatment of candidemia in patients without neutropenia. Candidemia Study Group and the National Institute. N Engl J Med 1994;331:1325–30. Anaissie EJ, Vartivarian SE, Abi-Said D, Uzun O, Pinczowski H, Kontoyiannis DP, et al. Fluconazole versus amphotericin B in the treatment of hematogenous candidiasis: a matched cohort study. Am J Med 1996;101:170 – 6.