A phase II trial of high-dose chemotherapy (HDCT) supported by hematopoietic stem-cell transplantation (HSCT) in germ-cell tumors (GCTs) patients failing cisplatin-based chemotherapy: the Multicentric TAXIF II study.

BACKGROUND: High-dose chemotherapy (HDCT) is an effective salvage treatment of germ-cell tumors (GCTs) patients. In the first salvage setting, 30%-70% of patients may achieve durable remissions. Even when HDCT is administered as subsequent salvage treatment, up to 20% of patients may still be definitively cured. However, patients with refractory/relapsed disease still have a very poor long-term prognosis, requiring earlier intervention of HDCT. PATIENTS AND METHODS: This phase II trial was addressed to nonrefractory patients failing Cisplatin-based chemotherapy. Inclusion criteria included seminomatous GCT in relapse after two lines of chemotherapy, nonseminomatous GCT in relapse after first or second lines, partial remission after first line, primary mediastinal GCT in first relapse. Patients received two cycles combining Epirubicin and Paclitaxel (Epi-Tax), followed by three consecutive HDCT, one using a Paclitaxel/Thiotepa (Thio-Tax) association and two using the 5-day Ifosfamide-Carboplatin-Etoposide regimen. The main objective was to determine the complete response rate. RESULTS: Forty-five patients were included between September 2004 and December 2007: 44 received the first HDCT cycle, 39 two HDCT cycles, 29 could receive the whole protocol. Sixteen patients did not receive the entire protocol, including eight (17.7%) for toxic side-effects. Two patients (4.4%) died of toxicities, and 17 (37.7%) of disease progression. With a median follow-up time of 26 months (range, 4-51), the final overall response rate was 48.8% (including a complete response rate of 15.5% and a partial response/negative serum markers rate of 26.6%) in an intent-to-treat analysis. The median progression-free survival (PFS) and overall survival (OS) times were 22 months [95% confidence interval (CI) 2-not reached] and 32 months (95% CI 4-49), respectively. The 2-year PFS was a plateau setup at 50% (95% CI 32-67) and the 2-year OS was 66% (95% CI 44-81). CONCLUSION: The TAXIF II protocol was effective in nonrefractory GCT patients failing Cisplatin-based chemotherapy. The toxic death rate remained acceptable in the field of HDCT regimens. TRIAL REGISTRATION NUMBER: NCT00231582.

Phase I study of high-dose topotecan with haematopoietic stem cell support in the treatment of ovarian carcinomas: the ITOV 01 protocol.

Topotecan has demonstrated activity in ovarian carcinomas. In order to increase the tumour response rate and to define the maximum tolerated dose (MTD) of topotecan, we decided to develop a high-dose phase I regimen supported by stem cell support. High-doses schedules using a 1-day single administration have MTDs of 10.5 (24 h continuous infusion (CI)) or 22.5 mg/m2 (30 min infusion). Five-day CI induces grade IV mucositis at high doses (MTD<12 mg/m2). We chose to administer topotecan in a 5-day schedule with a 30 min daily infusion. Patients were scheduled to receive one cycle of therapy. The first dose level was 4.0 mg/m2/day x 5 days. Limiting toxicities were defined as toxic death, grade IV non-haematopoietic or haematopoietic toxicity >6 weeks. From August 1998 to April 2002, 49 patients were included. Forty-three patients have completed one course and 15 have received two cycles. One patient treated at level 7 mg/m2/day died of sepsis. Median duration of grade IV neutropenia was 9 days. Two episodes of grade IV diarrhoea were observed at level 9.5 mg/m2/day. Pharmacokinetic data were linear within the dose range of 4-9.0 mg/m2/day. The MTD was reached at 9 mg/m2/day x 5 days.

High-dose sequential epirubicin and cyclophosphamide with peripheral blood stem cell support for advanced breast cancer: results of a phase II study.

The aim of this study was to evaluate the feasibility of a high-dose intensity and high-dose density multicycle epirubicin and cyclophosphamide regimen with peripheral blood stem cells (PBSC) and haematopoietic growth factor (G-CSF) support in advanced breast cancer patients. From August 1994 to September 1999, 56 breast cancer patients (8 stage IIIB and 48 stage IV) received 205 courses of cyclophosphamide 3 g x m(-2) and epirubicin 100 mg x m(-2) every 14 days. G-CSF 5 microg x kg(-1) x day(-1) was administered from day 3 to neutrophil recovery. 4 courses were planned. PBSC were collected after course 1, and reinfused after courses 3 and 4, with > or = 2 x 10(6) CD34+ PBSC x kg(-1) required for each reinfusion. 48 patients (86%) received all 4 planned courses. Early withdrawal was consecutive to infectious complications (n = 4), severe asthenia (n = 3), haemorrhagic cystitis (n = 1). A median number of 10.8 x 10(6) CD34+ PBSC x kg(-1) (range, 3-80) was harvested with 1 or 2 apheresis in 48 patients (94%). Median relative dose intensity was 91.3% (range, 72-102%). Grade 4 neutrophil toxicity was observed in 100% of patients. Febrile neutropenia was observed in 40% of courses (median duration 2 days). Red blood cells and platelets had to be transfused in 54% and 27% of courses, respectively. There were no toxic deaths. Objective response rate was 69% in stage IV patients (31/45 evaluable pts), with a 16% complete response rate. Their median progression-free and overall survivals were 22.5 and 37 months, respectively. This epirubicine-containing high-dose regimen appeared feasible, albeit with high toxicity. Time-related progression parameters exceed commonly reported ones. Controlled studies of upfront sequential high-dose chemotherapy are still needed to evaluate its real benefit.

Cytokines and vascular cell adhesion molecule-1 in the blood of patients undergoing HPC mobilization.

BACKGROUND: The mechanism of HPC mobilization in humans is unclear. In this study, the relationship between PBPC mobilization and blood levels of G-CSF, endogenous cytokines (IL-8, SCF, thrombopoietin [TPO]), and the vascular cell adhesion molecule-1 (VCAM-1) was analyzed in patients with malignancy who were undergoing a PBPC mobilization regimen. STUDY DESIGN AND METHODS: Fifty-four patients with multiple myeloma (MM) and 29 with breast cancer (BC) underwent a mobilization regimen combining conventional chemotherapy and G-CSF up to the last day of PBPC collection. The CD34+ cell count was determined on each day when leukapheresis was scheduled. Venous blood samples (n = 117) were drawn before apheresis for CD34+ cell count (flow cytometry) and cytokine (G-CSF, IL-8, SCF, TPO) and VCAM-1 measurements (ELISA). RESULTS: In multiple regression analysis, SCF was a significant determinant of CD34+ cell levels in BC patients (R = 0.50, p = 0.03) and of VCAM-1 levels in MM patients (R = 0.32, p = 0.02). SCF was negatively correlated with CD34+ cell count in patients with BC. SCF and VCAM-1 blood levels were correlated in MM and BC patients. CONCLUSION: SCF and VCAM-1 could play a role in PBPC mobilization in patients and could be useful measures by which to study patients undergoing a mobilization regimen.

Residual tumor cell contamination in peripheral blood stem cells collections of 117 breast cancer patients evaluated by immunocytochemical technique.

During the last years, high-dose chemotherapy and hematopoietic stem cell support have been thought to improve the treatment of poor-prognosis breast cancer. Nevertheless, the question remained as to whether the reinfusion of contaminating residual malignant cells could contribute to relapse. By using an immunocytochemical method, we have analyzed the tumor cell contamination of peripheral blood stem cells (PBSC) collected from advanced breast cancer patients. We studied 153 PBSC samples from 117 stage III and IV breast cancer patients and compared two screening methods-the usual microscopic observation and the automated cellular image analysis system (ACIS-assisted) screening. With manual observation, we found that 7 of 117 patients (5.9%) presented circulating epithelial tumor cells in 9 of 153 (5.8%) PBSC analyzed, whereas automated screening allowed positive detection in 15 of the same 117 patients (12.8%) and in 18 of the 153 PBSC (11.7%). No difference was found between presence or absence of circulating tumor cells and previous chemotherapy treatment (p = 0.5) or stage TNM (p = 0.13) in this group of poor-prognosis breast cancer. We did not find incidence of infusion of contaminated PBSC on overall survival or time to progression.

Functional G-CSF pathways in t(8;21) leukemic cells allow for differentiation induction and degradation of AML1-ETO.

INTRODUCTION: Efficacy of differentiating agents requires that their specific cellular targets are still expressed and functional in the leukemic cells. One hypothesis to target sensitive cells is to select leukemic clones which harbor disrupted transcription factors. CBFalpha and CBFbeta are core-binding proteins which have been identified as transcription regulators of hematopoietic genes and shown to be altered in numerous leukemias. In M2 AML, the t(8;21) translocation, CBFalpha (AML1) is altered and produced as the AML1-ETO fusion protein. The fusion protein blocks transcription and differentiation mediated by G-CSF. Interestingly, AML1-ETO leukemic cell lines are sensitive to numerous cytokines in vitro and can be induced to differentiate in the presence of G-CSF and PMA. MATERIALS AND METHODS: As in the APL differentiation model, primary culture provides a useful tool for therapeutic screening of differentiation inducers, we analysed the in vitro sensitivity of 10 fresh M2 AML t(8;21) leukemic samples to G-CSF and the functionality of G-CSF intracellular pathways. In vitro data were compared with in vivo data from four patients treated with rhG-CSF at the dosage of 5 microg/kg/day i.v. for two to three weeks before the initiation of AML induction chemotherapy and immunophenotypic analysis performed weekly to monitor in vivo differentiation. RESULTS: In vitro, an increase in CD34+ cells expressing differentiation antigens (CD11b, CD13 or CD15) was noted along with a decrease of immature CD34+/differentiation antigen negative cells. After two weeks of a daily rhG-CSF administration in vivo, a significant, albeit transient, decrease of blast count was achieved, concomitant with an increase in differentiated leukemic cells suggesting that in vivo differentiation occurs. Fresh t(8;21) leukemic cells possess functional G-CSF signaling pathways as normal activity and kinetics of STAT1 and STAT3 binding was observed. Furthermore, differentiation induction leads to a subsequent degradation of the AML1-ETO oncoprotein. CONCLUSION: The data presented here supports the claim that G-CSF can induce in vitro and in vivo differentiation of M2 AML t(8;21) cells.

Enrichment of peripheral blood CD34+ cells for transplantation using a fully automated immunomagnetic cell selection system and a novel octapeptide releasing agent.

Positive selection of CD34+ cells is being increasingly performed to support hematological reconstitution following high-dose and dose-intensive chemotherapy and to reduce the non-target cell content of transplants. The present study was designed to evaluate the performance of an immunomagnetic cell selection system, including comparison of enzyme and peptide releasing agents and of semi-automated and fully automated selection systems. A total of 74 immunomagnetic CD34+ cell selection procedures were performed involving 55 subjects, the majority of whom had hematologic malignancies. Median CD34+ cell purity with a newly developed specific octapeptide releasing agent (98.5%; 81.0-99.0%) was significantly higher (P = 0.002) than that with chymopapain (85.8%; 28.1-99.7%). No significant differences were observed between semi-automated and fully automated systems in CD34+ cell purity or yield or time to WBC or platelet recovery. Immunomagnetic selection was found to provide highly purified populations of CD34+ cells in sufficient numbers for use in transplantation procedures. CD34+ cell transplants supported rapid and reliable hematologic reconstitution. Use of a fully automated system markedly reduced the time and labor required for immunomagnetic selection, potentially affording more standardized and reproducible positive selection of CD34+ cells.

Collection of Ph-negative progenitor cells with granulocyte-colony stimulating factor in patients with chronic myeloid leukaemia who respond to recombinant alpha-interferon.

This study aimed to demonstrate that sufficient Ph-negative blood progenitors could be collected following administration of glycosylated rhG-CSF (lenograstim) to patients with Philadelphia chromosome (Ph)-positive chronic myeloid leukaemia (CML) who responded to recombinant alpha-interferon (alpha-IFN) (Ph-positive marrow metaphases < 35%). 23 patients received lenograstim (150 microg/m2) once daily for a median of 9 d. Peak circulating numbers of white blood cells (36.4 x 10(9)/l), CD34+ cells (24/ microl) and colony-forming unit-granulocyte-macrophage (CFU-GM; 1346.5/ml) occurred at a median of day 8, day 8 and day 7, respectively. Two to six (median three) leukaphereses (LK) were performed from days 5 to 12. The median number of mononuclear cells (MNC), CD34+ cells and CFU-GM collected per patient was 7.35 x 10(8/)kg, 2.72 x 10(6)/kg and 10.23 x 10(4)/kg, respectively. 22/23 patients had LK which contained either 10(4) CFU-GM/kg and/or 10(6) CD34+ cells/kg; 47LK (from 20/22 patients) were evaluable for cytogenetics. The median percentage of Ph-positive cells was 0, and 43/47 LK (91%) contained <35% Ph-positive cells; 25 (53%) were entirely negative. Sixteen of 20 evaluable patients had one or more LK with <35% Ph-positive cells, and 12 had at least one 100% Ph-negative LK. Mobilization and collection of Ph-negative cells were not influenced by the dose or duration of alpha-IFN administration before (or during) lenograstim administration or by the quality of cytogenetic response (complete v major) during lenograstim administration. No significant side-effects were observed. Thus, lenograstim administration can result in satisfactory Ph-negative blood progenitor cell collection. Autologous transplantation of such cells may be used when indicated.

Retroviral transfer and long-term expression of the adrenoleukodystrophy gene in human CD34+ cells.

Adrenoleukodystrophy (ALD) is a demyelinating disease of the central nervous system that results from a genetic deficiency of ALDP, an ABC protein involved in the transport of very long-chain fatty acids (VLCFAs). The cloning of the ALD gene and the positive effects of allogeneic bone marrow transplantation support the feasibility of a gene therapy approach. We report the retroviral transfer of the ALD cDNA to peripheral blood and bone marrow CD34+ cells from control donors and ALD patients. Prestimulation of these cells with cytokines, followed by infection with the M48-ALD retroviral vector, resulted in 20% transduction efficiency (4-40%) and expression of the vector-encoded ALDP in 20% of CD34+ cells (7.3-50%). Long-term culture (LTC) of transduced CD34+ cells from two ALD patients showed efficient transduction (24-28%) and stable expression (25-32%) of ALDP in derived clonogenic progenitors at 3 weeks of culture. The expression of ALDP in CFU cells derived from 5 and 6 weeks of LTC confirmed the effective transduction of LTC-initiating cells. Expression of ALDP was observed in CD68+ CFU-derived cells, suggesting that monocyte-macrophages, the target bone marrow cells in ALD, were produced from transduced progenitor cells. VL-CFA content was corrected in LTC and CFU-derived cells in proportion to the percentage of transduced cells, indicating that the vector-encoded ALDP was functional. Although not efficient yet to allow a clinical perspective, these results demonstrate the feasibility of ALD gene transfer into CD34+ cells of ALD patients.

[Autoimmune thrombopenic purpura. Clinical practices during diagnosis. A French survey and recommendations. French College of Hemotologists Evaluation Commission]. / Purpura thrombopénique auto-immun. Pratiques cliniques lors du diagnostic. Enquête française et recommandations.

THE SURVEY: To better ascertain diagnostic and treatment strategies used by physicians for idiopathic thrombocytopenic purpura, a questionnaire was addressed to 298 French hematologists and pediatricians. One hundred and ten responses were analyzed. DIAGNOSTIC APPROACH: Thrombocytopenia was determined on capillary blood samples by 50% of the physicians and platelet-associated antibody by 44%. Bone marrow aspirates were obtained more frequently in adults (85%) than in children (47%) and immune disorders were more often investigated in 95% of adults and in 68% of children. THERAPEUTICS: Treatment threshold was lower in children (20.10(9)/l) than in adults (50.10(9)/l). Corticoidsteroids was the treatment of choice in adults (98%) and children received either IV immunoglobulins (61%) and/or corticosteroids (63%); higher doses were used in children (> or = 2 mg/kg versus 1 mg/kg) for shorter periods (> or = 2 weeks versus > 3 weeks). Treatment failure was evaluated earlier in children (< 10 days) than in adults (> 20 days). RECOMMENDATIONS: 1. Diagnosis. Based on repeatedly low platelet counts and verification that only platelets are involved. Clinical examination is normal excepting rare cases of severe hemorrhage. Search for antiplatelet antibodies is non-contributive 2. Treatment Short-term corticosteroids, both in children and adults. Intravenous gammaglobulins should be limited to cases with signs of severe hemorrhage. 3. Consult the complete guidelines for each individual clinical situation.

Intensive chemotherapy with idarubicin, cytosine arabinoside, and granulocyte colony-stimulating factor (G-CSF) in patients with secondary and therapy-related acute myelogenous leukemia. Club de Réflexion en Hématologie.

Using a combination of intensive chemotherapy and G-CSF, we conducted a prospective trial designed to improve the complete remission (CR) rate in patients with AML evolving from a primary documented myelodysplastic syndrome (sAML) and therapy-related AML (tAML). Thirty-four patients (median age 61 years) with sAML (25 patients) or tAML (nine patients) entered the study. Induction course consisted of idarubicin (12 mg/m2 of body-surface area per day for 3 days) and intermediate-dose (ID) cytarabine in the 24 younger patients (1 g/m2 of body-surface area as a 2 h infusion every 12 h for 5 days) or standard-dose (SD) cytarabine in the 10 older patients (100 mg/m2 of body-surface area per day as a continuous infusion for 7 days), followed by G-CSF until neutrophil recovery or treatment failure. Nineteen patients (56%, 13/24 in the ID group and 6/10 in the SD group) achieved a CR (14/25 sAML and 5/9 tAML). Early death occurred in four patients, but four additional patients died in CR from treatment-related toxicity (overall toxic death rate 24%). Initial cytogenetics was available in 33 patients. The CR rate was significantly lower in patients with unfavorable cytogenetics compared to patients with intermediate cytogenetics (37% vs 79%). Median remission duration and overall survival were 3 and 9 months, respectively and not different between ID and SD patients. Although the treatment-related toxicity is high, a high CR rate can be obtained in these poor-risk AML patients with the use of intensive chemotherapy in combination with G-CSF, although the role of the latter is still to be proven. Results remain especially poor in patients with unfavorable cytogenetics. New approaches are needed to maintain remission in these high-risk AML patients.

Raynaud's phenomenon and digital necrosis induced by interferon-alpha.

We report two patients treated with interferon-alpha who developed Raynaud's phenomenon followed by multiple digital necrosis. Arteriography of the legs revealed diffuse distal narrowing. Histological examination of the resected tissue showed ischaemic necrosis within the dermis and subcutaneous tissue without vasculitis. Resolution of the symptoms was observed when interferon-alpha was withdrawn. Since other causes of Raynaud's phenomenon and digital necrosis were ruled out, a drug cause is likely. Raynaud's phenomenon should be looked for in patients treated with interferon-alpha. If present, immediate withdrawal of the drug is required.

Changes in microrheology of acute promyelocytic leukemia cells during all-trans retinoic acid (ATRA) differentiation therapy: a mechanism for ATRA-induced hyperleukocytosis?

According to French and European experience, hyperleukocytosis occurs during ATRA differentiation therapy in about 70% of de novo and 25% of relapsed APL cases. The most frequently suggested cause for this side-effect is an ATRA-induced proliferation of APL cells. However, no definite explanation for such a proliferative effect has been clearly established. Another mechanism directly related to the differentiation of marrow leukemic cells could be a change in their microrheology, allowing their release from the bone marrow and their transfer toward peripheral blood (PB) and tissues. Using a single cell aspiration assay into a glass restrictive channel, we measured APL cell viscosity values in five de novo APL patients. A deformability index (DI) was defined as the ratio of mean normal neutrophil viscosity x 100/mean APL cell viscosity. Results were the following: (1) at diagnosis, two patients had high marrow DI (96 and 250%) and three patients had low marrow DI (16, 17, and 40%); (2) when PB and marrow APL cells were simultaneously tested, PB APL cells display higher DI than marrow APL-cells; (3) the two patients with high initial marrow DI experienced an ATRA-induced hyperleukocytosis after only 1 day of treatment; (4) in the three patients with low initial marrow DI, the DI was increasing during ATRA therapy and hyperleukocytosis seemed to occur when a large amount of maturing APL cells reached a viscosity value similar to that of mature neutrophils. These results suggest that an asynchronism between rheological and morphological maturation in each APL cell might explain the occurrence of hyperleukocytosis in some patients during ATRA differentiation therapy.

Molecular analysis of clonality in Castleman's disease.

Castleman's disease (CD) is a rare atypical lymphoproliferative disorder that is morphologically and clinically heterogenous and is associated with a risk of developing malignant lymphoma. We report the clonality status of CD tissues in 34 patients, including 14 patients infected by the human immunodeficiency virus (HIV). Four patients presented a localized form and 30 presented a multicentric form. Two cases were associated with B-cell lymphoma, 3 cases with Hodgkin's disease, and 9 cases (8 HIV+) with Kaposi's sarcoma. Histologically, 8 cases were of the hyaline-vascular type and 26 were of the plasma cell or mixed types. The Ig and T-cell receptor (TCR) V(D)J rearrangements were analyzed using polymerase chain reaction and Southern blot. Clonal IgH rearrangements were detected in only 4 cases, ie, 2 associated with B-cell lymphoma, 1 with Hodgkin's disease, and 1 case without malignancy. A TCR gamma rearrangement of restricted junctional size was amplified in 1 HIV+ case. Finally, polyclonal VH-JH and V gamma-J gamma rearrangements were detected in the large majority of the cases, irrespective of pathologic subtypes, clinical forms, and HIV status. The lymphoid component in CD is therefore commonly reactive, and the rare occurrence of detectable monoclonal lymphoid contingents may be caused by secondary molecular events.

Unusual specific cutaneous lesions in myelodysplastic syndromes.

BACKGROUND: Early diagnosis of leukemia cutis in myelodysplastic syndrome (MDS) is important because these lesions can precede acute peripheral blood or bone marrow transformation. Leukemia cutis is usually easy to recognize, but atypical lesions are not well described. OBJECTIVE: Our purpose was to describe unusual specific lesions in MDS. METHODS: Data from patients with myeloid malignancies and leukemia cutis were reviewed. Only patients with MDS and cutaneous lesions different from typical tumors were included. Clinical features were analyzed. RESULTS: Four patients with MDS and unusual specific cutaneous lesions were found. They had ecchymoses, necrotic plaques or ulcers, and prurigo-like lesions. In three of four patients appearance of these skin lesions heralded or was concomitant with acute transformation. CONCLUSION: Specific cutaneous lesions can display unusual patterns. Early biopsy of cutaneous lesions in MDS is indicated.

Unexpected high incidence of intracranial subdural haematoma during intensive chemotherapy for acute myeloid leukaemia with a monoblastic component.

We report a high incidence of subacute, chronic and sometimes occult intracranial subdural haematoma (SDH) occurring during intensive chemotherapy for acute myeloid leukaemia (AML) with a monoblastic component. Between March 1990 and January 1993, 86 AML patients from our institution were randomized in the multicentric French AML 90 trial. Eight patients (9%) presented a grade > 2 haemorrhagic event, which was intracranial SDH in five of them. All these five SDH patients had hyperleucocytic AML4 or AML5 and had experienced at least one lumbar puncture (LP) before SDH diagnosis (with intrathecal chemotherapy in four cases). SDH diagnosis was assessed on a brain computed tomography scan which was performed 1-9 d after initial SDH symptoms (mainly mild headaches considered a result of prior LP). All these five patients recovered from this severe event after a specified therapy. SDH does not appear to be an uncommon complication of AML4 and AML5 therapy. Its incidence might be under-reported because of poor symptomatology. Lumbar punctures, known to cause exceptional SDH in nonleukaemic patients, might trigger these haemorrhagic events, eventually in combination with other predisposing factors such a haemostasis disorders or leukaemic CNS infiltration.

In vivo thrombin and plasmin activities in patients with acute promyelocytic leukemia (APL): effect of all-trans retinoic acid (ATRA) therapy.

APL-associated hemostasis disorders result from at least two distinct mechanisms due to the release of procoagulant activities and plasminogen activators from the leukemic cells. These two mechanisms (thrombin activation and plasmin activation) may cleave the fibrinogen molecule, but their respective roles in low fibrinogen levels and bleeding diathesis genesis remain in dispute. In vivo ATRA therapy induces a rapid correction of both low fibrinogen level and bleeding tendency, but no clear explanation of this beneficial effect has been proposed. We prospectively investigated 27 APL patients at presentation for diffuse intravascular coagulation (DIC) markers (prothrombin activation fragment and thrombin/antithrombin complexes) and plasmin-dependent primary fibrinogenolysis markers (alpha 2 plasmin inhibitor consumption +/- plasmin/alpha 2 plasmin inhibitor complexes). Fourteen of these patients were then serially studied during the first 2 weeks of ATRA therapy. Four of them, however, developed an hyperleukocytosis requiring additional chemotherapy before the end of the 2nd week. At presentation, low level of fibrinogen was clearly associated with alpha 2 plasmin inhibitor deficiency (p < 0.01), while DIC was equally present in fibrinogenopenic and non-fibrinogenopenic patients. Moreover, was observed a rapid simultaneous correction of low fibrinogen levels and plasmin activation markers in APL patients undergoing ATRA therapy (before day 5), but a more prolonged persistence of DIC markers (until day 14). Initial bleeding syndrome seemed more frequent in patients with initial low fibrinogen level. These data indicate that plasmin-dependent primary fibrinogenolysis is the major etiologic factor of low fibrinogen level in APL patients. In vivo differentiation ATRA therapy induces a rapid decrease in the plasmin activation and a normalization of fibrinogen level, while DIC may in vivo persist for several weeks. Prospective studies evaluating antifibrinolytic agents as therapy of APL-associated hemostasis disorders should be considered. Additionally, prophylactic heparin therapy might be useful after day 5 in patients undergoing ATRA therapy, since they present a prolonged procoagulant tendency.