A. Pifarré

Immunohistochemical analysis of anti-Hu-associated paraneoplastic encephalomyelitis

Abstract. The precise immune mechanisms of neuronal death in anti-Hu-associated paraneoplastic encephalomyelitis (PEM) are unclear. We performed an immunohistochemical study on postmortem brain tissue from 11 patients with anti-Hu-associated PEM to further characterize the immune reaction and to ascertain possible mechanisms of neuronal death. To analyze inflammatory infiltrates, antibodies against lymphocyte subpopulations (CD3, CD20, CD4, CD8), macrophage and activated microglia (CD68), major histocompatibility complex (MHC) classes I and II (HLA-ABC and HLA-DR), and the intercellular adhesion molecules (ICAM) -1 and -3 were used. Cell death mechanisms were defined using antibodies against the cytotoxic protein TIA-1, the C9neo component of complement, the Fas receptor (CD95) and its ligand, the apoptosis effector activated caspase-3, and the apoptosis inhibitor Bcl-2. A great number of T cells expressing the cytotoxic protein TIA-1 was observed, mainly in clusters around neurons. ICAM-1 immunoreactivity was increased in the neuropil and reactive astrocytes in areas of inflammation within the central nervous system and in satellite cells of pathological dorsal root ganglia surrounding apparently normal sensory neurons. By contrast, Fas, FasL, C9neo, and activated caspase-3 immunoreactivities were negative in pathological areas. Bcl-2 immunoreactivity was found in satellite cells, but not in sensory neurons of normal and pathological dorsal root ganglia. Our data point out to an induction of a cytotoxic, non-apoptotic, neuronal death in anti-Hu-associated PEM. The increased ICAM-1 immunoreactivity may favor the infiltration of lymphocytes in the pathological areas.

Detection of 14-3-3 brain protein in the cerebrospinal fluid of patients with paraneoplastic neurological disorders.

The detection of 14‐3‐3 protein in cerebrospinal fluid by immunoblotting is useful for the diagnosis of Creutzfeldt‐Jakob disease (CJD). We found 14‐3‐3 protein in 10 of 80 (12.5%) patients with paraneoplastic neurological disorders (PNDs), whose presenting symptoms may mimic those of CJD. In 47 of 48 CJD patients, the 14‐3‐3 protein was detected as a single band, and it was detected as a double band in 1 patient. The double‐band pattern was observed in 9 of the 10 14‐3‐3 protein–positive patients with PNDs. The 14‐3‐3 protein assay may be positive in PND patients, but the immunoblotting pattern distinguishes most PND samples from those of CJD.

Prognostic value of replication errors on chromosomes 2p and 3p in non-small-cell lung cancer

As chromosomes 2p and 3p are frequent targets for genomic instability in lung cancer, we have addressed whether alterations of simple (CA)n DNA repeats occur in non-small-cell lung cancer (NSCLC) at early stages. We have analysed by polymerase chain reaction (PCR) assay replication errors (RER) and loss of heterozygosity (LOH) at microsatellites mapped on chromosomes 2p and 3p in 64 paired tumour-normal DNA samples from consecutively resected stage I, II or IIIA NSCLC. DNA samples were also examined for K-ras and p53 gene mutations by PCR-single-stranded conformational polymorphism (PCR-SSCP) analysis and cyclic sequencing, as well as their relationship with clinical outcome. Forty-two of the 64 (66%) NSCLC patients showed RER at single or multiple loci. LOH was detected in 23 tumours (36%). Among patients with stage I disease, the 5-year survival rate was 80% in those whose tumours had no evidence of RER and 26% in those with RER (P = 0.005). No correlation was established between RER phenotype and LOH, K-ras or p53 mutations. RER remained a strong predictive factor (hazard ratio for death, 2.89; 95% confidence interval, 2.23-3.79; P = 0.002) after adjustment for all other evaluated factors, including p53, K-ras, LOH, histological type, tumour differentiation and TNM stage, suggesting that microsatellite instability on chromosomes 2p and 3p may play a role in NSCLC progression through a different pathway from the traditional tumour mechanisms of oncogene activation and/or tumour-suppressor gene inactivation.