Prognostic Relevance of NPM1 and FLT3 Mutations in Acute Myeloid Leukaemia, Longterm Follow-Up-A Single Center Experience.

The prognosis of acute myeloid leukemia depends on genetic aberrations, particularly NPM1 and FLT3-ITD mutations. The targeted drugs' availability has renewed interest in FLT3 mutations, but the impact of these genetic alterations using these treatments is yet to be confirmed. Our objective was to evaluate the results obtained with the intensified NILG-AML 01/00 protocol (ClinicalTrials.gov Identifier: NCT 00400673) in 171 unselected patients (median age, 54.5 years, range 15−74) carrying the FLT3 (ITD or TKD) and/or NPM1 mutations. The CR rate and 5-y survival were 88.3% and 58% +/− 4, respectively, significantly higher in the NPM1-mutated (CR 93.9%, p: 0.0001; survival 71% +/− 6, p: 0.0017, respectively). In isolated ITD patients, the CR was lower (66.7%, p: 0.0009), and the 3 years-relapse-free survival worse (24%, p: <0.0002). The presence of ITD, irrespective of the allelic ratio, or TKD mutation, did not significantly affect the survival or relapse-free survival among the NPM1-co-mutated patients. Our data indicate that a high dose of ARAC plus idarubicin consolidation exerts a strong anti-leukemic effect in NPM1-mutated patients both with the FLT3 wild-type and mutated AML, while in the NPM1 wild-type and FLT3-mutated, the therapeutic effect remains unsatisfactory. New strategies incorporating target therapy with second-generation inhibitors will improve these results and their addition to this aggressive chemotherapeutic program merits testing.

Validation of the "fitness criteria" for the treatment of older patients with acute myeloid leukemia: A multicenter study on a series of 699 patients by the Network Rete Ematologica Lombarda (REL).

OBJECTIVES: Treatment of older patients with acute myeloid leukemia (AML) is still controversial. To facilitate treatment decisions, the "fitness criteria" proposed by Ferrara et al. (Leukemia, 2013), including age > 75 years, performance status and comorbidities, were verified retrospectively in 699 patients with AML (419 de-novo, 280 secondary AML), diagnosed at 8 Hematological Centers (REL). METHODS: Patients were categorized in FIT to intensive chemotherapy (i-T) (292, 42.5%), UNFIT to i-T (289, 42.1%), or unfit even to non-intensive therapy (non i-T) (FRAIL) (105, 15.3%). Biological characteristics and treatment actually received by patients [i-T, 274 patients (39.2%); non i-T, 134 (19.2%), best-supportive care (BSC), 291 (41.6%)] were recorded. RESULTS: "Fitness criteria" were easily applicable in 98.1% of patients. Overall concordance between "fitness criteria" and treatment actually received by patients was high (79.4%), 76% in FIT, 82.7% in UNFIT and 80% in FRAIL patients. Fitness independently predicted survival (median survival: 10.9, 4.2 and 1.8 months in FIT, UNFIT and FRAIL patients, respectively; p = 0.000), as confirmed also by multivariate analysis. In FRAIL patients, survival with any treatment was no better than with BSC, in UNFIT non i-T was as effective as i-T and better than BSC, and in FIT patients i-T was better than non i-T or BSC. In addition, a non-adverse risk AML, an ECOG PS <2, and receiving any treatment other than BSC had a favorable effect on survival (p < 0.001). CONCLUSION: These simple "fitness criteria" applied at the time of diagnosis could facilitate, together with AML biologic risk evaluation, the choice of the most appropriate treatment intensity in older AML patients.

Postremission therapy with repeated courses of high-dose cytarabine, idarubicin, and limited autologous stem cell support achieves a very good long-term outcome in European leukemia net favorable and intermediate-risk acute myeloid leukemia.

Consolidation treatment in acute myeloid leukemia (AML) patients achieving complete remission (CR) is warranted. High-dose cytarabine (HDAC) is considered first choice in favorable risk and an option in intermediate-risk AML. However, its optimal dose and schedule, as well as the benefit of additional chemotherapy agents remain controversial. Herein, we report on the long-term outcome of consecutive unselected AML patients treated with repeated courses of HDAC, with the addition of idarubicin, followed by autologous peripheral blood stem cell (PBSC) support, in order to limit toxicity, according to Northern Italy Leukemia Group (NILG) AML-01/00 study (EUDRACT number 00400673). Among 338 patients consecutively diagnosed from 2001 to 2017 at our center, 148 with high-risk AML (adverse cytogenetic, isolated FLT3-internal tandem duplication mutation, refractory to first induction) were addressed to allogeneic stem cell transplant. All other cases, 186 patients (55%), median age 53 (range 19-75), were considered standard-risk and received the NILG AML-01/00 program. After achieving CR, patients were mobilized with cytarabine 8 g/sqm to collect autologous CD34+-PBSC and received three consolidation cycles with HDAC (20 g/sqm) plus idarubicin (20 mg/sqm) per cycle, followed by reinfusion of limited doses of CD34+ PBSC (1-2x106/kg). The program was completed by 160 (86%) patients. Toxicity was acceptable. Neutrophils recovered a median of 10 days. Treatment-related mortality was 3/160 (1.8%). After a median follow-up of 66.4 months, overall survival (OS) and relapse-free survival (RFS) at 5-years were 61.4% and 52.4%, respectively. Twenty-eight selected patients aged >65 had similar outcomes. According to European leukemia net-2010 classification, the OS and RFS at 5-years were 76.4% and 65% in favorable risk, without differences between molecular subgroups, 52.3% and 47.2% in Intermediate-I, 45.2% and 36.5% in Intermediate-II risk patients, respectively. In conclusion, consolidation including repeated courses of high dose cytarabine and idarubicin, with limited PBSC support, proved feasible and very effective in nonhigh risk patients. The incorporation of novel agents in its backbone may be tested to further improve patient's prognosis.

Transcription factors implicated in late megakaryopoiesis as markers of outcome after azacitidine and allogeneic stem cell transplantation in myelodysplastic syndrome.

The hypomethylating agent azacitidine (AZA) is used to treat higher-risk myelodysplastic syndromes (HR-MDS) and elderly patients with low-blast count acute myeloid leukemia (LBC-AML). Platelet recovery is an early predictor of AZA response. We prospectively studied the expression profile of transcription factors, critical for late megakaryopoiesis and changes in their expression after AZA treatment in patients with HR-MDS and LBC-AML enrolled in the BMT-AZA trial (EudraCT number 2010-019673-15). Twenty-five additional patients with low-risk (LR)-MDS were also studied. At the time of diagnosis, GATA2 mRNA levels were significantly higher in MDS as compared to controls, with increasing levels from LR- to HR-MDS/AML. RUNX1 expression was also significantly higher in MDS, as compared to controls, but no differences were found between LR- and HR-MDS. Looking at biomarkers of response, we found that patients AZA responsive had higher basal GATA1 and lower FLI1 expression, compared to those with stable or progressive disease after treatment. Univariate analysis showed that increased GATA2 mRNA expression was associated with a worse overall survival. Our findings suggest that high GATA2 expression is a poor prognostic marker for survival in patients with HR-MDS and LBC-AML treated with azacitidine. Moreover, GATA1 and FLI1 mRNA expression may predict response to AZA treatment.

Phase II trial with sequential clofarabine and cyclophosphamide for refractory and relapsed philadelphia-negative adult acute lymphoblastic leukemia. Results of the GIMEMA LAL 1610 protocol.

Clofarabine (CLO) and cyclophosphamide (CY) combinations were tested in late stage refractory/relapsed (R/R) acute lymphoblastic leukemia (ALL) with disappointing results and high-grade toxicity. We designed a sequential 5-day combination of CLO 40 mg/m2/d plus CY 400 mg/m2/d as first salvage for Philadelphia-negative ALL patients refractory or relapsed until 24 months from complete remission (CR). Primary endpoint was an overall response rate (ORR) ≥ 40%. Among 26 study patients (median age 40.5 years) ORR was 57.6% (CR with complete [n = 8] or incomplete [n = 7] hematologic recovery). Despite severe myelotoxicity, no dose-limiting toxicity suggested de-intensification of CLO schedule. With a median follow-up of 17.0 months, median and 1-year overall and disease-free survival were 6.5 months and 28.6%, and 3.7 months and 28.1%, respectively. This association was tolerable and more effective in patients younger than 40 years with B-precursor ALL, longer first CR, not previously transplanted and achieving CR with full hematological recovery.

Invasive pulmonary aspergillosis in acute leukemia: a still frequent condition with a negative impact on the overall treatment outcome.

We evaluated the impact of invasive pulmonary aspergillosis (IPA) on epidemiology and outcome in acute leukemia (AL), analyzing all acute myeloid (AML) and acute lymphoblastic leukemia (ALL) consecutively admitted to our Institution during a 5-year period of observation. Only AML patients received anti-mold prophylaxis. Among 175 AL patients (136 AML/39 ALL), possible and proven/probable IPA were diagnosed in 28 (16%). Frequency of IPA was similar in AML (16.2%) and in ALL (15.4%). Two-year overall survival (OS) was significantly affected by IPA (no IPA: 69.8% vs IPA: 31.7% p = .002). OS was similar in patients with proven/probable (28.2%) and possible IPA (36.4%) (p = .003 and .065, respectively). When censoring patients at transplant, IPA still affected 2-year survival (49.6% vs 79.2%, p = .02), but only proven/probable IPA was associated with lower survival (34.7%, p = .0003). IPA negatively impacts on long-term survival of leukemia patients; antifungal prophylaxis should be adopted also during induction in ALL and in AML beyond induction therapy.

The e13a2 BCR-ABL transcript negatively affects sustained deep molecular response and the achievement of treatment-free remission in patients with chronic myeloid leukemia who receive tyrosine kinase inhibitors.

BACKGROUND: Stopping tyrosine kinase inhibitor (TKI) treatment has become a realistic and safe objective for patients who have chronic myeloid leukemia (CML). Both a sustained deep molecular response (sDMR) and the lack of a molecular recurrence after TKI discontinuation are required to reach a durable treatment-free remission (TFR). METHODS: The potential predictive role of BCR-ABL transcripts in attaining an sDMR and a TFR was analyzed in a strictly consecutive, unselected series of 194 patients who were diagnosed and treated with TKIs at the authors' center. RESULTS: Of 173 fully evaluable patients, 67 (38.7%) had the e13a2 transcript, and 106 (61.3%) had the e14a2 transcript. Complete cytogenetic and major molecular remissions were not affected, whereas the achievement of both a DMR (P = .008) and an sDMR (P = .004) was favored significantly in patients who had the e14a2 transcript. After a median of 68 months, the sDMR rate was 39.6% in those with the e14a2 transcript and 19.4% in those with the e13a2 transcript. In addition to transcript type, both the early achievement of a molecular response and starting treatment with a second-generation TKI positively affected the attainment of an sDMR in multivariate analysis. The use of a second-generation TKI as frontline treatment increased the sDMR rate in both transcript types. However, in patients who had the e13a2 transcript, the probability of attaining an sDMR was 37% after 60 months and did not increase further despite continuing therapy. Among 51 of 60 patients who attained an sDMR after discontinuing TKIs, 24 experienced a molecular relapse, but all regained molecular remission after resuming TKI treatment. Again, transcript type influenced TFR maintenance (P = .005), because only 2 patients (3%) with the e13a2 transcript enjoyed a durable TFR compared with 25 (23.5%) of those with the e14a2 transcript. CONCLUSIONS: The e13a2 transcript hinders the achievement of deep responses and the possibility of stopping TKI treatment in patients with CML.

Molecular remission at the end of treatment is a necessary goal for a good outcome in ELN favorable-risk acute myeloid leukemia: a real-life analysis on 201 patients by the Rete Ematologica Lombarda network.

Favorable acute myeloid leukemia (AML) patients (pts.) demonstrate a relatively good outcome with standard induction; thus, pts. are generally not addressed to allogeneic transplant in first remission. However, it is not clear if also in a real-life setting, the outcome is homogeneous in the different favorable molecular groups and which are the parameters significantly associated to an increased relapse risk, useful to suggest the need of an intensified approach. In order to clarify this point, we collected clinical data on consecutive unselected AML pts. assigned to favorable category (modified ELN 2010 due to the inclusion of double-mutated CEBPA-positive cases), diagnosed and treated in six centers of the Italian network Rete Ematologica Lombarda (REL) from 2007 to 2015. We assessed response (CR, mCR), relapse rate (CIR), and outcome (OS, DFS) after first-line treatment. A total of 201 pts. was studied and the analysis was performed globally and in each molecular group: t(8;21)(q22;q22)/RUNX1-RUNX1T1 (30 pts., 14.9%), inv. (16)(p13q22) or t(16;16)(p13q22)/CBFB-MIH11 (35 pts., 17.4%), normal karyotype and mutated NPM1 and negative FLT3-ITD (116 pts., 57.7%) or double-mutated CEBPA (CEBPAdm) (20 pts., 10%). Complete remission (CR) was obtained in 188 pts. (93.5%), molecular CR (mCR) in 114 (67.5%); After a median follow-up of 2.4 years, cumulative incidence of relapse (CIR) was documented in 78 of 188 responding pts. (41%) after a median time of 11.3 months. CIR was higher in the CBFB-MIH11 group, in pts. achieving only a hematological response without mCR (72.1 vs 28.1%, p < 0.001), in older pts. and it resulted independently associated with a lower median cytarabine cumulative dose (CCD). Median OS was not reached: after 5 years it was 66.3%, and median DFS was 5.3 years, both without difference among groups. Molecular CR reached at any time, during or after the end of first-line treatment, was significantly associated with better DFS, and in particular, mCR assessed at the end of treatment was confirmed in multivariate analysis as an independent prognostic factor both for DFS and OS. In conclusion, the present study confirms in a real-life context the overall good prognosis of favorable-risk AML; the achievement of any molecular negativity during first-line treatment, particularly when assessed at the end of treatment, is associated with lower relapse and better survival. Increasing age at diagnosis has a negative prognostic impact, while CCD higher than 18 g/sqm is associated with better outcome.

Baseline factors associated with response to ruxolitinib: an independent study on 408 patients with myelofibrosis.

In patients with Myelofibrosis (MF) treated with ruxolitinib (RUX), the response is unpredictable at therapy start. We retrospectively evaluated the impact of clinical/laboratory factors on responses in 408 patients treated with RUX according to prescribing obligations in 18 Italian Hematology Centers. At 6 months, 114 out of 327 (34.9%) evaluable patients achieved a spleen response. By multivariable Cox proportional hazard regression model, pre-treatment factors negatively correlating with spleen response were: high/intermediate-2 IPSS risk (p=0.024), large splenomegaly (p=0.017), transfusion dependency (p=0.022), platelet count <200×109/l (p=0.028), and a time-interval between MF diagnosis and RUX start >2 years (p=0.048). Also, patients treated with higher (≥10 mg BID) average RUX doses in the first 12 weeks achieved higher response rates (p=0.019). After adjustment for IPSS risk, patients in spleen response at 6 months showed only a trend for better survival compared to non-responders. At 6 months, symptoms response was achieved by 85.5% of 344 evaluable patients; only a higher (>20) Total Symptom Score significantly correlated with lower probability of response (p<0.001). Increased disease severity, a delay in RUX start and titrated doses <10 mg BID were associated with patients achievinglower response rates. An early treatment and higher RUX doses may achieve better therapeutic results.

Postremission sequential monitoring of minimal residual disease by WT1 Q-PCR and multiparametric flow cytometry assessment predicts relapse and may help to address risk-adapted therapy in acute myeloid leukemia patients.

Risk stratification in acute myeloid leukemia (AML) patients using prognostic parameters at diagnosis is effective, but may be significantly improved by the use of on treatment parameters which better define the actual sensitivity to therapy in the single patient. Minimal residual disease (MRD) monitoring has been demonstrated crucial for the identification of AML patients at high risk of relapse, but the best method and timing of MRD detection are still discussed. Thus, we retrospectively analyzed 104 newly diagnosed AML patients, consecutively treated and monitored by quantitative polymerase chain reactions (Q-PCR) on WT1 and by multiparametric flow cytometry (MFC) on leukemia-associated immunophenotypes (LAIPs) at baseline, after induction, after 1st consolidation and after 1st intensification. By multivariate analysis, the factors independently associated with adverse relapse-free survival (RFS) were: bone marrow (BM)-WT1 ≥ 121/10(4) ABL copies (P = 0.02) and LAIP ≥ 0.2% (P = 0.0001) (after 1st consolidation) (RFS at the median follow up of 12.5 months: 51% vs. 82% [P < 0.0001] and 57% vs. 81%, respectively [P = 0.0003]) and PB-WT1 ≥ 16/10(4) ABL copies (P = 0.0001) (after 1st intensification) (RFS 43% vs. 95% [P < 0.0001]) Our data confirm the benefits of sequential MRD monitoring with both Q-PCR and MFC. If confirmed by further prospective trials, they may significantly improve the possibility of a risk-adapted, postinduction therapy of AML.

Lenalidomide in patients with red blood cell transfusion-dependent myelodysplastic syndrome and del(5q): a single-centre "real-world" experience.

"Real life" data are needed to complement published trials on the efficacy of lenalidomide in patients with myelodysplastic syndrome (MDS) and del(5q) and on the risk of inducing acute myeloid leukemia (AML) progression. Here, we present results of lenalidomide treatment in a consecutive, population-based series of 21 red blood cell (RBC) transfusion-dependent elderly patients with multiple comorbidities. Of 18 evaluable patients (median follow-up: 22 months), 17 achieved an erythroid hematologic response (HI-E) and 16 an RBC transfusion independence. Cytogenetic response (CyR) rate was 80%, median overall survival was 48 months (range 3-164), and 5-year leukemia-free survival was 84%. Three patients progressed to AML; one, with baseline TP53 mutation, achieved HI-E, partial CyR, and did not progress to AML. Eighteen patients experienced hematological adverse events. Overall, lenalidomide was very effective and well tolerated even in unselected elderly patients with multiple comorbidities and did not appear to increase the risk of AML.

Randomized phase III trial of retinoic acid and arsenic trioxide versus retinoic acid and chemotherapy in patients with acute promyelocytic leukemia: health-related quality-of-life outcomes.

PURPOSE: A randomized clinical trial compared efficacy and toxicity of standard all-trans-retinoic acid (ATRA) plus chemotherapy versus ATRA plus arsenic trioxide in patients with newly diagnosed, low- or intermediate-risk acute promyelocytic leukemia (APL). Here, we report health-related quality-of-life (HRQOL) results. PATIENTS AND METHODS: HRQOL was a secondary end point of this trial. The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 was used to assess HRQOL at end of induction and after consolidation therapy. All analyses were based on 156 patients who received at least one dose of treatment, with groups defined according to randomly assigned treatment. Primary analysis was performed, estimating mean HRQOL score over time and differences between treatment arms using a linear mixed model. RESULTS: Overall, 162 patients age 18 to 70 years were enrolled. Of these, 150 and 142 patients were evaluable for HRQOL after induction therapy and third consolidation course, respectively. Overall compliance with HRQOL forms was 80.1%. The largest difference, favoring patients treated with ATRA plus arsenic trioxide, was found for fatigue severity (mean score difference, -9.3; 95% CI, -17.8 to -0.7; P = .034) at end of induction therapy. This difference was also clinically relevant. HRQOL differences between treatment arms at end of consolidation showed that for several scales, differences between treatment arms were marginal. CONCLUSION: Overall, current HRQOL findings further support the use of ATRA plus arsenic trioxide as preferred first-line treatment in patients with low- or intermediate-risk APL.

Relapsing bloodstream infections during treatment of acute leukemia.

Acute leukemia (AL) patients may experience more than one episode of bloodstream infection (BSI) caused by the same pathogen during the entire chemotherapy program. In order to identify factors influencing BSI recurrence (R-BSI) during subsequent phases of treatment, we analyzed all BSIs occurring to consecutively treated AL patients during a period of active epidemiologic surveillance at our institution between 2004 and 2011. Two hundred and fifty BSIs were observed in 138 patients receiving more than 1 cycle of chemotherapy. BSI due to the same pathogen recurred in 39/138 (28.3 %) patients. Gram-negative rods (GNRs) accounted for 59.6 % and Gram-positive cocci (GPCs) for 34.4 % of BSI. Four pathogens were involved in R-BSI: Escherichia coli, Pseudomonas aeruginosa, coagulase-negative staphylococci, and Streptococcus viridans. GNRs were significantly more frequent among R-BSI compared to non-relapsing BSI (nR-BSI) [69/94 (73.4 %) vs 70/156 (50.6 %), p < 0.0001]; in particular, E. coli accounted for 67 % of R-BSI vs 32.1 % of nR-BSI (p < 0.0001). Receiving more than four chemotherapy courses and having an extended spectrum ß-lactamase (ESBL)-producing E. coli BSI at any time of treatment were significantly associated to R-BSI. A trend toward a higher mortality among R-BSI patients in comparison with nR-BSI was observed (17.9 and 7.1 %, respectively, p = 0.12). Among AL patients, R-BSI is a frequent phenomenon, which may contribute to the shift of epidemiology toward GNR and to a higher mortality. This should significantly impact the strategies of antibiotic prophylaxis and treatment in patients with AL.

Retinoic acid and arsenic trioxide for acute promyelocytic leukemia.

BACKGROUND: All-trans retinoic acid (ATRA) with chemotherapy is the standard of care for acute promyelocytic leukemia (APL), resulting in cure rates exceeding 80%. Pilot studies of treatment with arsenic trioxide with or without ATRA have shown high efficacy and reduced hematologic toxicity. METHODS: We conducted a phase 3, multicenter trial comparing ATRA plus chemotherapy with ATRA plus arsenic trioxide in patients with APL classified as low-to-intermediate risk (white-cell count, ≤10×10(9) per liter). Patients were randomly assigned to receive either ATRA plus arsenic trioxide for induction and consolidation therapy or standard ATRA-idarubicin induction therapy followed by three cycles of consolidation therapy with ATRA plus chemotherapy and maintenance therapy with low-dose chemotherapy and ATRA. The study was designed as a noninferiority trial to show that the difference between the rates of event-free survival at 2 years in the two groups was not greater than 5%. RESULTS: Complete remission was achieved in all 77 patients in the ATRA-arsenic trioxide group who could be evaluated (100%) and in 75 of 79 patients in the ATRA-chemotherapy group (95%) (P=0.12). The median follow-up was 34.4 months. Two-year event-free survival rates were 97% in the ATRA-arsenic trioxide group and 86% in the ATRA-chemotherapy group (95% confidence interval for the difference, 2 to 22 percentage points; P<0.001 for noninferiority and P=0.02 for superiority of ATRA-arsenic trioxide). Overall survival was also better with ATRA-arsenic trioxide (P=0.02). As compared with ATRA-chemotherapy, ATRA-arsenic trioxide was associated with less hematologic toxicity and fewer infections but with more hepatic toxicity. CONCLUSIONS: ATRA plus arsenic trioxide is at least not inferior and may be superior to ATRA plus chemotherapy in the treatment of patients with low-to-intermediate-risk APL. (Funded by Associazione Italiana contro le Leucemie and others; ClinicalTrials.gov number, NCT00482833.).

A randomized comparison of caspofungin versus antifungal prophylaxis according to investigator policy in acute leukaemia patients undergoing induction chemotherapy (PROFIL-C study).

BACKGROUND: Invasive fungal infections (IFIs) are considered a major problem among patients undergoing acute leukaemia (AL) induction treatment. PROphylaxis of Fungal invasive Infections in Leukaemia-Caspofungin (PROFIL-C) is a multicentre study aiming to assess the comparative yield of using caspofungin versus standard policy (SP) regimens and the overall impact of IFI in routine clinical care conditions. METHODS: All AL patients receiving IFI prophylaxis according to local SP were prospectively included in the study by Northern Italy Leukaemia Group (NILG) centres. To allow the comparison of caspofungin versus SP regimens as prophylaxis strategies, caspofungin treatment was assigned via a centralized randomized procedure. The study was registered at http://www.clinicaltrial.gov (NCT00501098). RESULTS: Over a 2 year period, 175 patients were included. The overall incidence of IFI was 32/175 (18.3%) [10/175 (5.7%) probable/proven and 22/175 (12.6%) possible], with no statistically significant differences between caspofungin-based versus SP-based regimens [overall: 15/93 (16.1%) versus 17/82 (20.7%), relative risk (RR) 0.78, 95% confidence interval (CI) 0.42-1.46; probable/proven: 7/93 (7.5%) versus 3/82 (3.7%), RR 2.06, 95% CI 0.55-7.7; possible: 8/93 (8.6%) versus 14/82 (17.1%), RR 0.5, 95% CI 0.22-1.14]. Only one IFI-related death was recorded (10%). CONCLUSIONS: The incidence and mortality of IFI were lower than expected in this strictly sequential cohort representative of the routine care in the NILG network. The efficacy and safety of caspofungin were similar to other prophylactic regimens.