The actin-related proteins.

A family of proteins has been discovered over the past three years whose members have clear sequence homology to actin but are distinguished from actin by their structural and functional diversity. The ranks of this family, whose members are known as the actin-related proteins (arps), are expanding rapidly. Arps are but one branch of a larger superfamily which includes the actins, hsp/hsc70s, sugar kinases and several cell cycle proteins from bacteria. The existence of the superfamily has been inferred from tertiary structural data. In the case of the arps, their identification and classification has been based upon primary structural data. Placing the arps in a functional context is proving a slower process, although genetic and biochemical analyses are converging in several cases. In the past year, different arps have been linked to functions mediated by actin filaments (arp2 and arp3), microtubules (arp1) and the structural elements of chromatin (arp4 and arp6).

Transcutaneous immunization: a human vaccine delivery strategy using a patch.

Transcutaneous immunization, a topical vaccine application, combines the advantages of needle-free delivery while targeting the immunologically rich milieu of the skin. In animal studies, this simple technique induces robust systemic and mucosal antibodies against vaccine antigens. Here, we demonstrate safe application of a patch containing heat-labile enterotoxin (LT, derived from Escherichia coli) to humans, resulting in robust LT-antibody responses. These findings indicate that TCI is feasible for human immunization, and suggest that TCI may enhance efficacy as well as improve vaccine delivery.

Protection of macaques against vaginal transmission of a pathogenic HIV-1/SIV chimeric virus by passive infusion of neutralizing antibodies.

The development of the human immunodeficiency virus-1 (HIV-1)/simian immunodeficiency virus (SIV) chimeric virus macaque model (SHIV) permits the in vivo evaluation of anti-HIV-1 envelope glycoprotein immune responses. Using this model, others, and we have shown that passively infused antibody can protect against an intravenous challenge. However, HIV-1 is most often transmitted across mucosal surfaces and the intravenous challenge model may not accurately predict the role of antibody in protection against mucosal exposure. After controlling the macaque estrous cycle with progesterone, anti-HIV-1 neutralizing monoclonal antibodies 2F5 and 2G12, and HIV immune globulin were tested. Whereas all five control monkeys displayed high plasma viremia and rapid CD4 cell decline, 14 antibody-treated macaques were either completely protected against infection or against pathogenic manifestations of SHIV-infection. Infusion of all three antibodies together provided the greatest amount of protection, but a single monoclonal antibody, with modest virus neutralizing activity, was also protective. Compared with our previous intravenous challenge study with the same virus and antibodies, the data indicated that greater protection was achieved after vaginal challenge. This study demonstrates that antibodies can affect transmission and subsequent disease course after vaginal SHIV-challenge; the data begin to define the type of antibody response that could play a role in protection against mucosal transmission of HIV-1.

Human skin Langerhans cells are targets of dengue virus infection.

Dengue virus (DV), an arthropod-borne flavivirus, causes a febrile illness for which there is no antiviral treatment and no vaccine. Macrophages are important in dengue pathogenesis; however, the initial target cell for DV infection remains unknown. As DV is introduced into human skin by mosquitoes of the genus Aedes, we undertook experiments to determine whether human dendritic cells (DCs) were permissive for the growth of DV. Initial experiments demonstrated that blood-derived DCs were 10-fold more permissive for DV infection than were monocytes or macrophages. We confirmed this with human skin DCs (Langerhans cells and dermal/interstitial DCs). Using cadaveric human skin explants, we exposed skin DCs to DV ex vivo. Of the human leukocyte antigen DR-positive DCs that migrated from the skin, emigrants from both dermis and epidermis, 60-80% expressed DV antigens. These observations were supported by histologic findings from the skin rash of a human subject who received an attenuated tetravalent dengue vaccine. Immunohistochemistry of the skin showed CD1a-positive DCs double-labeled with an antibody against DV envelope glycoprotein. These data demonstrate that human skin DCs are permissive for DV infection, and provide a potential mechanism for the transmission of DV into human skin.

Endogenous peptides of a soluble major histocompatibility complex class I molecule, H-2Lds: sequence motif, quantitative binding, and molecular modeling of the complex.

To gain insight into the rules that govern the binding of endogenous and viral peptides to a given major histocompatibility complex (MHC) class I molecule, we characterized the amino acid sequences of a set of self peptides bound by a soluble analogue of murine H-2Ld, H-2Lds. We tested corresponding synthetic peptides quantitatively for binding in several different assays, and built three-dimensional computer models of eight peptide/H-2Lds complexes, based on the crystallographic structure of the human HLA-B27/peptide complex. Comparison of primary and tertiary structures of bound self and antigenic peptides revealed that residues 2 and 9 were not only restricted in sequence and tolerant of conservative substitutions, but were spatially constrained in the three-dimensional models. The degree of sequence variability of specific residues in MHC-restricted peptides reflected the lack of structural constraint on those amino acids. Thus, amino acid residues that define a peptide motif represent side chains required or preferred for a close fit with the MHC class I heavy chain.

Insulin increases H2O2-induced pancreatic beta cell death.

Insulin resistance results, in part, from impaired insulin signaling in insulin target tissues. Consequently, increased levels of insulin are necessary to control plasma glucose levels. The effects of elevated insulin levels on pancreatic beta (ß) cell function, however, are unclear. In this study, we investigated the possibility that insulin may influence survival of pancreatic ß cells. Studies were conducted on RINm, RINm5F and Min-6 pancreatic ß-cells. Cell death was induced by treatment with H(2)O(2), and was estimated by measurements of LDH levels, viability assay (Cell-Titer Blue), propidium iodide staining and FACS analysis, and mitochondrial membrane potential (JC-1). In addition, levels of cleaved caspase-3 and caspase activity were determined. Treatment with H(2)O(2) increased cell death; this effect was increased by simultaneous treatment of cells with insulin. Insulin treatment alone caused a slight increase in cell death. Inhibition of caspase-3 reduced the effect of insulin to increase H(2)O(2)-induced cell death. Insulin increased ROS production by pancreatic ß cells and increased the effect of H(2)O(2). These effects were increased by inhibition of IR signaling, indicative of an effect independent of the IR cascade. We conclude that elevated levels of insulin may act to exacerbate cell death induced by H(2)O(2) and, perhaps, other inducers of apoptosis.

Exercise peripheral oxygen saturation (SpO2) accurately reflects arterial oxygen saturation (SaO2) and predicts mortality in systemic sclerosis.

BACKGROUND: Measures of oxygenation have not been assessed for prognostic significance in systemic sclerosis-related interstitial lung disease (SSc-ILD). METHODS: 83 subjects with SSc-ILD performed a maximal cardiopulmonary exercise test with an arterial line. The agreement between peripheral oxygen saturation (SpO2) and arterial oxygen saturation (SaO2) was examined and survival differences between subgroups of subjects stratified on SpO2 were analysed. Cox proportional hazards analyses were used to examine the prognostic capabilities of SpO2. RESULTS: At maximal exercise the mean (SD) difference between SpO2 and SaO2 was 2.98 (2.98) and only 15 subjects had a difference of >4 points. The survival of subjects with SSc-ILD whose maximum exercise SpO2 (Spo(2)max) fell below 89% or whose SpO2max fell >4 points from baseline was worse than subjects in comparator groups (log rank p = 0.01 and 0.01, respectively). The hazard of death during the median 7.1 years of follow-up was 2.4 times greater for subjects whose SpO2max fell below 89% (hazard ratio 2.4, 95% CI 1.1 to 4.9, p = 0.02) or whose SpO2max fell >4 points from baseline (hazard ratio 2.4, 95% CI 1.1 to 5.0, p = 0.02). CONCLUSION: In patients with SSc-ILD, SpO2 is an adequate reflection of SaO2 and radial arterial lines need not be inserted during cardiopulmonary exercise tests in these patients. Given the ease of measurement and its prognostic value, SpO2 should be considered as a meaningful clinical and research outcome in patients with SSc-ILD.

Replication of HIV-1 in dendritic cell-derived syncytia at the mucosal surface of the adenoid.

Human immunodeficiency virus-type 1 (HIV-1) replicates actively in infected individuals, yet cells with intracellular depots of viral protein are observed only infrequently. Many cells expressing the HIV-1 Gag protein were detected at the surface of the nasopharyngeal tonsil or adenoid. This infected mucosal surface contained T cells and dendritic cells, two cell types that together support HIV-1 replication in culture. The infected cells were multinucleated syncytia and expressed the S100 and p55 dendritic cell markers. Eleven of the 13 specimens analyzed were from donors who did not have symptoms of acquired immunodeficiency syndrome (AIDS). The interaction of dendritic cells and T cells in mucosa may support HIV-1 replication, even in subclinical stages of infection.

Phase II trial of oral vorinostat (suberoylanilide hydroxamic acid) in relapsed diffuse large-B-cell lymphoma.

BACKGROUND: Vorinostat has demonstrated activity in refractory cutaneous T-cell lymphoma. In a phase I trial, an encouraging activity in diffuse large-B-cell lymphoma (DLBCL) was noted. PATIENTS AND METHODS: We carried out a phase II trial (NCT00097929) of oral vorinostat 300 mg b.i.d. (14 days/3 weeks or 3 days/week) in patients with measurable, relapsed DLBCL who had received two or more systemic therapies. Response rate and duration (DOR), time to progression (TTP) and safety were assessed. RESULTS: Eighteen patients were enrolled (median age: 66 years; median prior therapies: 2). Seven received 300 mg b.i.d. 14 days/3 weeks, but four had grade 3 or 4 toxicity (dose-limiting toxicity, DLT). The schedule was amended to 300 mg b.i.d. 3 days/week), and none had DLT. One achieved a complete response (TtR = 85 days; DOR =or >468 days) and one had stable disease (301 days). Sixteen discontinued for progressive disease; median TTP was 44 days. Median number of cycles was 2 (1 to >19). Common drug-related adverse experiences (AEs; mostly grade 1/2) were diarrhea, fatigue, nausea, anemia and vomiting. Three patients had dose reduction; none discontinued for drug-related AEs. Drug-related AE >or=grade 3 included thrombocytopenia (16.7%) and asthenia (11.1%). CONCLUSION: Vorinostat was well tolerated at 300 mg b.i.d. 3 days/week or 200 mg b.i.d. 14 days/3 weeks but had limited activity against relapsed DLBCL.

Potential mechanisms of human natural killer cell expansion in vivo during low-dose IL-2 therapy.

The continuous, in vivo infusion of low-dose IL-2 selectively expands the absolute number of human natural killer (NK) cells after 4-6 weeks of therapy. The mechanism responsible for this expansion is unknown and was examined in this study. NK cells cultured at low concentrations of IL-2, comparable to those found during in vivo therapy, proliferate for 6 days and then exit the cell cycle. However, NK cells in vivo did not traverse the S/G(2)/M phase of the cell cycle during low-dose IL-2 therapy. Low concentrations of IL-2 delay programmed cell death of NK cells but have the same effect on resting T cells that do not expand in vivo. When CD34(+) bone marrow hematopoietic progenitor cells are cultured for 21 days with low concentrations of IL-2, they differentiate into CD56(+)CD3(-) NK cells, not T cells. Thus, the selective expansion of human NK cells during continuous in vivo infusion of low-dose IL-2 likely results from enhanced NK-cell differentiation from bone marrow progenitors, combined with an IL-2-dependent delay in NK-cell death, rather than proliferation of mature NK cells in the periphery.

Variability and accuracy in mammographic interpretation using the American College of Radiology Breast Imaging Reporting and Data System.

BACKGROUND: Several studies, which were limited by their small sample size and selection of difficult cases for review, have reported substantial variability among radiologists in interpretation of mammographic examinations. We have determined, in the largest study to date, intraobserver and interobserver agreement in interpreting screening mammography and accuracy of mammography by use of the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS). METHODS: The mammographic examinations were randomly selected on the basis of original mammographic interpretation and cancer outcome from 71,713 screening examinations performed by the Mobile Mammography Screening Program of the University of California, San Francisco, during the period from April 1985 through February 1995. The final sample included 786 abnormal examinations with no cancer detected, 267 abnormal examinations with cancer detected, and 1563 normal examinations. Films were read separately by two radiologists according to BI-RADS. Cancer status was determined by contacting women's physicians and by linkage to the regional Surveillance, Epidemiology, and End Results Program. RESULTS: There was moderate agreement between radiologists in reporting the presence of a finding when cancer was present (kappa = 0.54) and substantial agreement when cancer was not present (kappa = 0.62). Agreement was moderate in assigning one of the five assessment categories but was statistically significantly lower when cancer was present relative to when cancer was not present (kappa = 0.46 versus 0.56; two-sided P = .02). Agreement for reporting the presence of a finding and mammographic assessment was two-fold more likely for examinations with less dense breasts. Agreement was higher on repeat readings by the same radiologists than between radiologists. The sensitivity of mammography was lower with BI-RADS than with the original system for mammographic interpretation, but the positive predictive value of mammography was higher. CONCLUSION: Considerable variability in interpretation of mammographic examinations exists; this variability and the accuracy of mammography are neither improved nor diminished with use of BI-RADS.

Clinical pharmacology of oral all-trans retinoic acid in patients with acute promyelocytic leukemia.

All-trans retinoic acid (RA) induces leukemic cell differentiation and complete remission in a high proportion of patients with acute promyelocytic leukemia (APL). However, remissions induced by all-trans RA tend to be brief, and relapses are associated with resistance to further treatment in vivo, although the leukemic cells appear to retain sensitivity to the cytodifferentiating effects of all-trans RA in vitro. The clinical pharmacology of all-trans RA was examined in 13 patients with APL. The drug was administered at a constant dose of 45 mg/m2/day, given as a single dose on the first day of therapy and in two divided doses thereafter. Plasma and urinary concentrations of the parent drug and metabolites were quantitated by reverse-phase high-performance liquid chromatography and, where required, by a combination of normal-phase liquid chromatography/negative chemical ionization mass spectrometry. In patients with APL, basal levels of endogenous retinol and natural retinoids were within the normal range. Peak plasma levels of all-trans RA (347 +/- 266 ng/ml, mean +/- SD) were reached 1-2 h after drug ingestion and decayed in a monoexponential fashion with a half-life of 0.8 +/- 0.1 h. The only drug metabolite detected in plasma or urine was 4-oxo-all-trans RA (present in urine as the glucuronide conjugate). This metabolite accounted for less than 10% of the circulating drug in plasma, and its cumulative urinary excretion accounted for less than 1% of the administered dose. The drug was not found in cerebrospinal fluid. Continued oral administration of all-trans RA was associated with a significant decrease in both the plasma peak levels and the area under the concentration-time curve (P = 0.01 and 0.004, respectively) when measured after 2-6 weeks of treatment. We previously reported that a decrease in plasma area under the concentration-time curve was highly correlated with clinical relapse. Observations in a subset of patients in this study suggested that, in fact, the major decrease occurred early, within the first 7 days of treatment. These changes were associated with a 10-fold increase in urinary excretion of 4-oxo-all-trans RA glucuronide, suggesting that the accelerated clearance from plasma was associated with increased drug catabolism. The rapid disappearance may explain early relapse from remissions induced by all-trans RA; clinical "resistance" to all-trans RA may either wholly or in part result from an inability to sustain effective plasma concentrations of all-trans RA during continuous treatment.(ABSTRACT TRUNCATED AT 400 WORDS)

Molecular rearrangement of the ALL-1 gene in acute myeloid leukemia without cytogenetic evidence of 11q23 chromosomal translocations.

Translocations which involve chromosome band 11q23 are frequently found in infants and adults with acute myeloid leukemia (AML) or acute lymphoblastic leukemia. We previously cloned a gene called ALL-1 which spans the 11q23 breakpoint and is rearranged in most cases of leukemia with 11q23 abnormalities. In the present report, we have investigated the occurrence of ALL-1 rearrangement in cases of AML without cytogenetic evidence of 11q23 abnormalities. We detected molecular rearrangements of the ALL-1 gene in 3 of 4 patients with de novo AML and trisomy 11 as a sole chromosomal abnormality. Furthermore, we found DNA rearrangements of ALL-1 in 2 of 19 patients with de novo AML and normal cytogenetics. We conclude that molecular rearrangement of ALL-1 often can be detected in de novo AML, despite the absence of cytogenetic abnormalities involving 11q23.

Utility of magnetic resonance imaging in the management of breast cancer: evidence for improved preoperative staging.

PURPOSE: The staging and treatment for breast cancer are changing; there is an increase in the incidence of ductal carcinoma-in-situ, the use of fine-needle aspiration and stereotactic biopsy for diagnosis, and the use of neoadjuvant chemotherapy. Thus, there is a need for a tool to assess more precisely the extent of cancer in the breast before surgery. To better plan surgical and chemotherapeutic interventions, we evaluated high-resolution magnetic resonance imaging (MRI) as such a tool. PATIENTS AND METHODS: Fifty-seven patients with 58 cases of breast cancer were evaluated preoperatively with MRI using a technique called the triple-acquisition rapid gradient echo technique to maximize anatomic detail. Imaging results were compared with mammography and subsequent pathology results. RESULTS: Magnetic resonance imaging correctly identified residual or primary cancer in 55 of 58 cases and accurately predicted the extent of the cancer in 54 of 58 cases. The anatomic extent was more accurately defined with MRI compared with mammography (98% v 55%). Magnetic resonance imaging added the greatest value in cases of multifocal disease. CONCLUSION: By applying MRI selectively to patients with a known diagnosis of cancer and focusing on defining the extent of malignant lesions, we were able to obtain clear and accurate anatomic information. Our results suggest that MRI could provide very valuable information for preoperative planning and single-stage resection in breast cancer. Based on preliminary data from our series, MRI would be valuable as a staging tool in the preoperative setting even if the cost is in the range of $1,300 to $2,000. It is already significantly less than the target cost, so it is reasonable to refine this technique for clinical use to help plan the most appropriate surgical intervention and possibly reduce costs as well. A careful prospective study is warranted to validate our findings.

Phase I and II trial of five-day infused 5-fluorouracil and radiation in advanced cancer of the head and neck.

Eighteen patients with advanced epithelial cancers of the head and neck region were studied for their tolerance and response to combined cycles of 120-hour infused 5-fluorouracil (5-FU) and external-beam radiation therapy. 5-FU infusions were given under conditions where radiosensitization would be expected at the higher infusion doses. Coincident radiation treatments were given as four sequential daily fractions of 250 rad each administered during days 1 through 4 of each five-day infusion cycle. The patients were rested for at least nine days after each cycle or longer until toxicity was resolved. The regimen was then repeated in each patient for a total of five treatment cycles. Thereafter therapy was consolidated, usually by boost radiation without drug. In sequential patient subsets the infusion load was progressively escalated in a phase I format. The complete response rate for stage IV patients was 75% with survival benefit compared to prior results. 5-FU dose-dependent combined modality loco-regional toxicity was demonstrated without significant enhancement of systemic toxicity of any form; 5-FU dose-dependent enhanced responsiveness and survival benefit is also suggested. Further scheduling and response studies of 5-FU under radiosensitizing conditions appear warranted.

Randomized, multicenter, open-label study of pegfilgrastim compared with daily filgrastim after chemotherapy for lymphoma.

PURPOSE: The primary objective was to assess the duration of grade 4 neutropenia (neutrophil count < 0.5 x 10(9)/L) after one cycle of chemotherapy with etoposide, methylprednisolone, cisplatin, and cytarabine in patients randomly assigned to receive one dose of pegfilgrastim or daily filgrastim after chemotherapy. Febrile neutropenia, neutrophil profiles, time to neutrophil recovery, pharmacokinetics, and safety were also assessed. PATIENTS AND METHODS: An open-label, randomized, phase II study was designed to compare the effects of a single subcutaneous injection of pegfilgrastim (sustained-duration filgrastim) 100 micro g/kg per chemotherapy cycle (n = 33) with daily subcutaneous injections of filgrastim 5 micro g/kg (n = 33) in patients receiving salvage chemotherapy for relapsed or refractory Hodgkin's or non-Hodgkin's lymphoma. RESULTS: The incidence of grade 4 neutropenia in the pegfilgrastim and filgrastim groups was 69% and 68%, respectively. In addition, the mean duration of grade 4 neutropenia was similar in both groups (2.8 and 2.4 days, respectively). The results for the two groups were also not significantly different for febrile neutropenia, neutrophil profile, time to neutrophil recovery, or toxicity profile. A single subcutaneous injection of pegfilgrastim 100 micro g/kg produced a sustained serum concentration relative to daily subcutaneous injections of filgrastim. Filgrastim-treated patients received a median of 11 injections per cycle. CONCLUSION: Pegfilgrastim was safe and well tolerated in this patient population. A single injection of pegfilgrastim per chemotherapy cycle provided neutrophil support with safety and efficacy similar to that provided by daily injections of filgrastim. Once-per-cycle administration of pegfilgrastim simplifies the management of neutropenia and may have important clinical benefits for patients and healthcare providers.

United States multicenter study of arsenic trioxide in relapsed acute promyelocytic leukemia.

PURPOSE: To determine the safety and efficacy of arsenic trioxide (ATO) in patients with relapsed acute promyelocytic leukemia (APL). PATIENTS AND METHODS: Forty patients experiencing first (n = 21) or > or = second (n = 19) relapse were treated with daily infusions of ATO to a maximum of 60 doses or until all leukemic cells in bone marrow were eliminated. Patients who achieved a complete remission (CR) were offered one consolidation course of ATO that began 3 to 4 weeks later. Patients who remained in CR were eligible to receive further cycles of ATO therapy on a maintenance study. RESULTS: Thirty-four patients (85%) achieved a CR. Thirty-one patients (91%) with CRs had posttreatment cytogenetic tests negative for t(15;17). Eighty-six percent of the patients who were assessable by reverse transcriptase polymerase chain reaction converted from positive to negative for the promyelocytic leukemia/retinoic acid receptor-alpha transcript by the completion of their consolidation therapy. Thirty-two patients received consolidation therapy, and 18 received additional ATO as maintenance. Eleven patients underwent allogeneic (n = 8) or autologous (n = 3) transplant after ATO treatment. The 18-month overall and relapse-free survival (RFS) estimates were 66% and 56%, respectively. Twenty patients (50%) had leukocytosis (> 10,000 WBC/microL) during induction therapy. Ten patients developed signs or symptoms suggestive of the APL syndrome and were effectively treated with dexamethasone. Electrocardiographic QT prolongation was common (63%). One patient had an absolute QT interval of > 500 msec and had an asymptomatic 7-beat run of torsades de pointe. Two patients died during induction, neither from drug-related causes. CONCLUSION: This study establishes ATO as a highly effective therapy for patients with relapsed APL.

Identification of a divergent actin-related protein in Drosophila.

The actin-related proteins (ARPs) have primary sequence homology to actin, have no homology to other proteins and, unlike the conventional actins, are clearly divergent. We have identified an ARP in Drosophila that has approximately 30% amino acid identity to most actins, making it the most divergent yet reported. It is also quite divergent from all other ARP sequences. When the Drosophila ARP is aligned with actin it contains sequence insertions, as is the case with all other ARPs. The unique location of the insertions, as well as its overall divergence, indicates it may represent a new isotype. Only one gene was detected by hybridization to both genomic DNA and polytene chromosomes; the location of the gene is 13E on the X chromosome. A transcript of 1350 bases was detected at all stages of development. This transcript was relatively abundant during early embryogenesis, decreasing during the later stages of embryogenesis and increasing again in larvae and adults.

Cloning of a secretory gelsolin from Drosophila melanogaster.

Gelsolin is an actin-binding protein with the abilities to sever and cap the barbed end of actin filaments and to promote the assembly of monomeric actin. It has been identified in vertebrates both as a cytoplasmic protein and as a protein secreted into the blood plasma. Here we report the nucleic acid sequence of the full-length complementary DNA for a secretory form of gelsolin from Drosophila. The deduced amino acid sequence of 790 residues (M(r) = 87,669) contains a predicted signal peptide of 20 amino acid residues. Comparison of the Drosophila gelsolin sequence with other members of the gelsolin family of actin-binding proteins reveals the characteristic segmental repeat structure found in this class of proteins. A 42% identity is observed between Drosophila secretory gelsolin and human plasma gelsolin when their primary structures are compared. Northern blots resolve a single 3000 base message in third instar Drosophila larvae, a message that appears to be encoded by a single gene located at 82A, B on the right arm of the third chromosome.

Treatment of acute promyelocytic leukemia with all-trans retinoic acid: an update of the New York experience.

In an ongoing study, we treated 79 patients with a molecular diagnosis of acute promyelocytic leukemia (APL) using all-trans retinoic acid (RA) for remission induction. Newly diagnosed patients received cytotoxic chemotherapy for consolidation, and previously treated patients received extended all-trans RA therapy, or a radionuclide-conjugated monoclonal antibody as post-remission treatment. Unlike studies in Europe, full-dose chemotherapy was not given during induction for patients who developed leukocytosis. Overall, 43 of 49 newly diagnosed patients (88%) and 25 of 30 previously treated patients (83%) achieved complete remission. We did not encounter de novo resistance to all-trans RA in any patient who was positive for PML/RAR-alpha rearrangements by reverse transcription polymerase chain reaction analysis. Ten patients died during induction from intracranial or pulmonary hemorrhage (six patients) or the 'retinoic acid syndrome' (four patients). The use of leukapheresis or low-dose chemotherapy (hydroxyurea or cytosine arabinoside) for drug-induced leukocytosis did not decrease early mortality. Compared to historical controls, early mortality was not affected by treatment with all-trans RA; however, both relapse-free and overall survival were significantly increased. Maintenance therapy with all-trans RA was associated with short remission duration, and relapses while taking the drug were universally associated with resistance to further retinoid treatment. We conclude that the use of all-trans RA for remission induction, with or without full-dose chemotherapy, has significantly increased the survival of patients with APL. While early mortality has not yet been reduced, the avoidance of full-dose chemotherapy during induction has significantly reduced early morbidity. The major outstanding clinical issue is the development of strategies that maximize safety in high-risk patients for whom intracranial hemorrhage remains the major cause of death.