T. P. T. Chan

Central nervous system involvement in non-Hodgkin's lymphoma

Fifty-eight Hong Kong Chinese patients with CNS lymphoma were reviewed (primary seven, secondary 51). The incidence of secondary CNS lymphoma in patients with non-Hodgkins lymphoma was estimated to be 9.4%. The Working Formulation separated subtypes which had a special propensity to involve the CNS. Significant proportions of our patients with secondary CNS lymphoma had other features which were known to be associated with a high risk of CNS disease including stage IV (48/51, 91.4%), bone marrow (26/51, 50.9%), peripheral blood (7.51, 13.7%), nasal (7/51, 13.7%), orbital (3/51, 5.9%), testicular (2/51, 3.9%) and bulky retroperitoneal (6/51, 11.8%) disease. 82% of patients with secondary CNS lymphoma had concurrent systemic disease and a further 12% had systemic relapse shortly afterward. CNS lymphoma is associated with poor prognosis and only 29% and 14% of the patients with primary and secondary CNS lymphoma respectively survived beyond 1 year. Patients responding to therapy had significantly better survival. 69.9% of the deaths were related to progressive systemic disease.

EGFR Array: Uses in the Detection of Plasma EGFR Mutations in Non–Small Cell Lung Cancer Patients

Introduction: We aim to develop a simple and sensitive array-based method for the detection of epidermal growth factor receptor (EGFR) gene mutations in the plasma of non–small-cell lung cancer patients and determine its use in the follow-up of those on tyrosine-kinase inhibitor (TKI) therapy. Method: DNA from 100 &mgr;l of plasma was amplified in the presence of peptide nucleic acid clamp to provide single-stranded template for the allele-specific arrayed primer extension reaction, incorporating cyanine-5-deoxycytidine triphosphate in the newly synthesized strands. The fluorescent product was visualized by laser at 670 nm. Results: Eleven different types of EGFR TKI drug-sensitive mutants (SM) were identified in plasma-DNA from 46 of 51 patients. Five patients carried only wild-type sequence. Plasma-DNA finding was concordant in 36 of 37 cases with tumor-sequencing data. This method could detect as little as 62.5 copies of mutant L858R; 125 copies of E709K + G719A or 625 copies of del 746–750 in the presence of 100,000 copies of wild-type EGFR. In 21 patients on longitudinal follow-up for up to 18 months, SM was found in all initial plasma samples, except for three samples collected after recent chemotherapy. Nine of 16 patients (56%) who responded to TKI had undetectable plasma EGFR mutant. SM was present concurrently with drug-resistant mutant in 44% of patients with disease progression while on TKI, the remaining 56% might have other mechanisms of resistance. Conclusion: The EGFR array provides a sensitive, inexpensive, and robust method for monitoring non–small-cell lung cancer patients’ response to TKI, and obviates the need of repeated lung biopsy.

Use of the oral chelator deferiprone in the treatment of iron overload in patients with Hb H disease

Seventeen non‐transfusion‐dependent Chinese haemoglobin H (Hb H) disease patients (age 29–76 years) with serum ferritin >900 μg/l were treated with deferiprone for up to 18 months. One patient withdrew and data from 16 patients were analysed. Sixteen other Hb H patients with ferritin <900 μg/l, matched for age and genotype, acted as controls. Treatment was well tolerated except for mild arthralgia. Serum ferritin fell with treatment, reaching significance at 6 and 18 months (from 1492·3 ± 901·4 to 519·4 ± 405·4 μg/l at 18 months, P = 0·0008). Nine of 16 patients had levels below 397 μg/l before 18 months. Serum ferritin remained stable 6 months after stopping treatment. In contrast, there was no change in ferritin levels in the control group. Magnetic resonance imaging was used for measurement of liver iron content. Spin echo T1‐signal intensity ratio (T1‐SIR) and gradient echo T2‐signal intensity ratio (T2‐SIR) increased with treatment. T2‐SIR rose from 0·17 ± 0·08 pretreatment to 0·58 ± 0·50 at 2 years (P = 0·0055). Improvement occurred in 12 of 16 patients, reaching normal in three patients. Using echocardiography, peak early diastolic : late diastolic blood flow (E/A) remained unchanged with treatment, but isovolumic relaxation time (IVRT) was prolonged at 2 years indicating mild impairment of diastolic function. All systolic function parameters were normal. A longer treatment period is desirable to demonstrate improvement in cardiac function.

Multiple Xbal polymorphisms for carrier detection and prenatal diagnosis of haemophilia A

Three Xbal restriction fragment length polymorphisms (RFLPs) can be detected using the factor VIII‐intron 22 probe (p482.6) in a Xbal‐KpnI double digest of genomic DNA. The Xbal (A) site had been reported by Wion et al (1986) to be in intron 22. while the two additional sites. Xbal (B) and Xbal (C), are shown here to be X‐linked and close to the Xbal (A) site. The frequencies of heterozygosity for these three sites are 0.49. 0.18 and 0.30 respectively. In 75 females the observed heterozygosity rate for the Xbal (A) site is 0.41 and this increased to 0.57 with the two additional sites. Care should be exercised when interpreting the Xbal RFLPs, since the 1.4 kb Xbal/Kpnl fragment and the 4.8 kb Xbal fragment are associated with both positive Xbal (A) and Xbal (B) sites. By the combined use of the multiple Xbal polymorphisms with the Bell site in intron 18, the carrier detection rate would increase to 67%. Four prenatal diagnoses had been performed using the multiple Xbal polymorphisms.

Molecular defects in haemophilia B: detection by direct restriction enzyme analysis

The common restriction fragment length polymorphisms (RFLPs) associated with the FIX gene: 5’BamH I, Dde I, BamH I (2), Taq 1 and 3’Hha I were absent or of low incidence in Southern Chinese and are therefore not useful for linkage analysis. No deletion was detected amongst seven consecutive unrelated haemophilia B patients, but one had an insertion of a 15 kb Pvu II fragment containing exon d. Using an alternate strategy of polymerase chain reaction (PCR) amplification and direct sequencing, the molecular defect in the other six patients was defined. The four novel mutations characterized were: nucleotide (nt) 6410 G→C (Gly12→Ala): nt 31261 Δ T (stop codon 31 bp downstream); nt 31260 C→G (Thr380→Ser) and nt 31122 C→A (Ala34→Asp). Two patients had the same mutation at nt 6365, G→A (Arg−4→Gln), identical to one previously described in other ethnic groups, suggesting that this is a hotspot for mutation. Each of the mutations was found to affect an enzyme recognition site and could thus be identified by direct visualization of abnormal restriction fragments in amplified genomic DNA. This allows rapid and accurate DNA diagnosis of haemophilia B in an ethnic group which otherwise shows little or no polymorphism for the common RFLP sites.

Molecular characterization of haemophilia A in southern Chinese

41 unrelated southern Chinese haemophilia A patients were studied. The 5′ promoter region, all 26 exons, their immediate 5′ and 3′ flanking splice junctions and the 3′ untranslated region of the FVIII gene were amplified (including 16 different segments of exon 14) using GC‐clamped primers. The GC‐clamped PCR products were screened by denaturing gradient gel electrophoresis (DGGE) and fragments showing an abnormal migration pattern were sequenced. 17 mutations were identified, of which four were splicing defects, involving the first 1–6 nucleotide (nt) in the intervening sequences (IVS), six missense mutations, three nonsense mutations and four frameshift mutations. 14 other patients carried the type 1 inversion, affecting the distal copy of the F8A gene at the telomere of the X chromosome and the same gene in intron 22 of the FVIII gene. All the mothers studied (12/14) were carriers of the inversion. Two of these patients with inversion also have a co‐existing missense mutation. In most cases the clinical severity of the disease corresponds to the genotype.