[Intravenous amphotericin B as prevention of deep mycoses in allogeneic bone marrow transplantation]

BACKGROUND

To evaluate the efficacy of i.v. amphotericin B (AmB) as prophylaxis of deep mycosis (DM) in allogenic bone marrow transplantation (BMT).

METHODS

From July 1991 to May 1993, 45 consecutive patients treated by allogenic BMT with no previous history of systemic mycosis and with normal renal function were administered prophylactic AmB at a dose of 0.5 mg/kg/48 h from day + 1 BMT until hemoperipheral recovery (group A). These were compared with an historic control group made up of 45 consecutive patients submitted to BMT from January 1990 to June 1991 who did not receive prophylactic AmB (group B). During the neutropenic phase all the patients remained in isolation units with laminar flow of filtered air and were administered oral non absorbable antibiotic therapy and diet of low bacterial content. The incidence of DM and the dose of AmB administered during the first 120 days post BMT were evaluated.

RESULTS

In the first 30 days following BMT 3 (7%) cases of DM were observed in group A and 3 (7%) in group B. Four (9%) additional cases were found from days 30 to 120 in group A and 3 (7%) in group B. In 3 (7%) patients of the group which received prophylaxis and in 4 (9%) of the control group Candida spp. was isolated. In 3 (7%) patients from group A and 1 (2%) patient from group B the infection was due to Aspergillus. Although the patients from group A received therapeutic AmB less frequently (78% vs 91%) and later (13 [SD +/- 5.9] vs 9.2 [SD +/- 4.6] days) than those of group B (p < 0.002) the mean dose of AmB per patient treated was similar in both groups (11.3 [SD +/- 8.8] vs 11.8 [SD +/- 7] mg/kg).

CONCLUSIONS

The prophylactic use of systemic amphotericin during the neutropenic phase of bone marrow transplantation does not reduce either the incidence of deep mycosis or the mean dose of amphotericin administered.

A sensitive enzyme immunoassay for angiotensin II in serum

A sensitive and specific enzyme immunoassay for measuring angiotensin II (AII) has been developed as a convenient alternative to a radioimmunoassay. An antiserum to AII was prepared using AII conjugated by carbodi-imide to rabbit serum albumin, and coated on to microwell plates. The labelled antigen was prepared from AII and horseradish peroxidase using the periodate method. This enzyme immunoassay was a simple two-step procedure: 0.1 ml of AII-extracted plasma was incubated for 1 h at 37 degrees C; and 1 ml of labeled AII was incubated for 1 h at 37 degrees C. Bound horseradish peroxidase activity was then determined using o-phenylenediamine as chromogen by measuring the absorbance at 492 nm. The lower detection limit of the assay was 3.5 pmol l-1. Between- and within-assay RSD values were 8.8-18.3% and 6.9-17%, respectively, for concentrations of 10-40 pmol l-1. The accuracy of the assay, determined by recovery and linearity experiments, was 89-106% for recovery and 91-126% for parallelism. The results obtained by the present ELISA method were well correlated with those obtained by an established radioimmunoassay (n = 10, r = 0.96, intercept = 0.9 and slope = 1.02). This assay is easy to perform, rapid and does not require radioisotopes; thus it could be widely applied in clinical laboratories.

Decrements in heart rate variability during memory search

Heart-rate variability (HRV) was recorded from subjects as they rested or performed two memory-search tasks with different memory-set sizes. Results indicated that all the estimates of HRV were considerably greater during resting periods than during the task but only the mid-frequency band (0.1 Hz component) reflected the changes in memory-set size by a more pronounced decrement with the higher load. Subjective mental-effort ratings collected for the task periods also reflected the different memory loadings but did not correlate with HRV measures.

Expression of HLA class I and II antigens in bronchogenic carcinomas: its relationship to cellular DNA content and clinical-pathological parameters

We studied the presence of HLA class I antigens in 115 samples of bronchogenic carcinomas (66 frozen and 49 formalin-fixed and paraffin-embedded specimens) by the immunophosphatase alkaline and immunoperoxidase methods with antibodies against major histocompatibility complex antigens. We also studied HLA class II antigens on the 66 frozen tumor samples. Nonneoplastic lung tissue was also analyzed for purposes of comparison. Pneumocytes and epithelial respiratory cells expressed HLA class I and II antigens. The expression of class I antigens was totally lost in 29 tumors (25%). The defect in HLA gene expression affected both heavy chain and beta 2-microglobulin, as demonstrated by the null reactivity with specific antibodies. In 2 cases of 66 studied in cryostatic section, the selective loss of A locus was observed, and in three cases selective loss of B locus was detected. The expression of class I antigens was compared with clinical-pathological parameters such as histological type, degree of differentiation, and tumor stage, as well as tumoral ploidy. The absence of expression of HLA class I molecules was significantly associated with poorly differentiated and undifferentiated tumors (P less than 0.0001) and with aneuploid tumors (P less than 0.001), suggesting that some lung tumors may escape immune surveillance and become biologically more aggressive. Class II antigens were expressed in 13 cases of 66 studied (18%) in frozen specimens, and a clear relationship was observed with well-differentiated tumors (P less than 0.05).

Altered HLA class I expression in non-small cell lung cancer is independent of c-myc activation

We studied the expression of major histocompatibility complex class I antigens in 59 bronchogenic carcinomas, as well as in pneumocytes and epithelial respiratory cells distant from the tumor. We observed in all cases that normal lung tissue expressed major histocompatibility complex class I antigens, while this expression was completely lost in 16 tumors (27%). The defect in HLA gene expression affected both heavy chain and beta 2-microglobulin, as demonstrated by the null reactivity with the monoclonal antibodies GRH1, W6/32, and HC10. Selective underexpression was detected in 1 tumor for HLA-A locus antigens and in 3 tumors for HLA-B locus antigens. Southern blot analyses of major histocompatibility complex class I genes were performed in 20 tumor tissue specimens and 6 cell lines. No class I gene rearrangements were detected using HLA coding and locus specific noncoding probes. We also used the Southern blot method to investigate the possible relationship between c-myc amplification and HLA class I antigens in non-small cell lung cancers and detected no apparent amplification in 20 tumor tissue specimens (5 negative for HLA class I antigens) and 6 cell lines (3 with decreased expression). Northern blot analysis revealed no relationship between c-myc mRNA levels and specific mRNA for HLA-A and HLA-B antigens in cell lines with imbalanced HLA-A or HLA-B expression.

Molecular analysis of MHC-class-I alterations in human tumor cell lines

Molecular characterization of HLA-class-I expression was investigated in human tumor cell lines at the protein and mRNA levels using locus-specific monoclonal antibodies (MAbs) and probes. Some cell lines exhibited a differential expression of HLA-A and HLA-B products and also showed differences in the inducibility of HLA-class-I genes by gamma-IFN. Thus, gamma-IFN stimulation induced predominantly HLA-B mRNA in the HeP-2 cell line, which showed imbalances in basal levels of HLA-A and HLA-B expression. This unequal inducibility of HLA genes may imply that locus-specific regulatory mechanisms are involved in the expression of individual HLA products. The specific mechanism controlling the differential expression of HLA subsets appears to be independent of c-myc activity. Northern blot analysis found no relationship between c-myc mRNA levels and specific mRNA for HLA-A and HLA-B antigens.

K-ras mutations (codon 12) are not involved in down-regulation of MHC class-I genes in colon carcinomas

Fifty-eight colorectal carcinomas were studied for HLA class-I antigen expression and for the presence of point mutations in codons 12 and 61 of the K-ras gene. Eight carcinomas were completely negative for class I by the APAAP technique. Analyses using the polymerase chain reaction (PCR) method, together with selective hybridization using mutation-specific synthetic oligonucleotides, demonstrated K-ras mutations in 14 cases (24.1%), all of them in codon 12. None of the mutations corresponded to the negative cases for class-I HLA antigen expression. We did not observe any correlation between K-ras mutations and the extent of tumor differentiation.

Presence of HPV 16 sequences in laryngeal carcinomas

Human papillomavirus types HPV 16 and HPV 11 DNA sequences were analyzed in normal and neoplastic tissues of the larynx, using the technique of polymerase chain reaction (PCR). An amplified region of E6 ORF was hybridized with 3' end-labelled oligonucleotide probe. Twenty six out of 48 (54%) squamous-cell carcinomas, and 3 out of 3 verrucous-cell carcinomas hybridized with HPV 16 DNA sequences, whereas we did not detect HPV 11 sequences. HPV 16 DNA sequences were also found in normal, autologous mucosa and lymphnode metastases, although these were absent in other tissues analyzed. HPV-16-positive tumors were most frequently poorly differentiated squamous-cell carcinomas.

MHC class I and II antigens on gastric carcinomas and autologous mucosa

The expression of HLA class I and II antigens was analysed in 30 primary gastric carcinomas, 27 autologous lymph node metastases and 25 autologous gastric mucosae. We used an immune alkaline phosphatase technique on cryostatic sections and mAbs directed against HLA class I monomorphic determinants, HLA-B locus-specific products and HLA-DR, -DP and -DQ molecules. In addition HLA class I genes were analysed in tumour tissue and compared by Southern blots with the RFLP from autologous mucosa using locus-specific HLA probes. Finally the infiltrating mononuclear cells were studied on gastric tumours and adjacent mucosa with mAbs defining CD4, CD8 and CD11b differentiation antigens. The results obtained showed that three out of 27 primary gastric carcinomas completely lack HLA-ABC antigens (10%). In addition, two primary tumours presented a variable expression. The remaining 22 tumours presented a homogeneous positive HLA class I expression. Interestingly, when the autologous mucosa was analysed, only 12 out 25 specimens were homogeneously stained with mAbs against HLA class I antigens, suggesting that this tissue may lack the expression of HLA antigens before becoming malignant. Indeed, the majority of the gastric carcinomas studied presented a higher HLA-ABC antigenic expression than autologous mucosa. Finally, the HLA expression observed in the primary tumour was similar to that observed in autologous metastases. As a second part of the study we have found a direct relationship between the expression of HLA-DR antigens in mucosa and the intensity of inflammatory infiltration. This relationship was not maintained in the tumour tissue. In the mucosa the CD4-positive T cell was the predominant lymphocyte, while it was CD8 in the HLA-DR-positive tumours. Finally the RFLP of class I genes did not show any differences in any of the cases when compared with autologous mucosa. We included in these studies DNAs from HLA class I-negative tumours, HLA positive and HLA-B-negative ones.