Background: The protocols of WHO network laboratories facilitated development of rapid diagnosis for SARS coronavirus (CoV) using reverse transcription (RT)-PCR assays detecting the pol gene sequence. However, several reports have shown that conventional and real-time PCR assays were very specific for SARS CoV but lack sensitivity depending on the assay, specimen, and time course of disease.

Study design: Single or multiple nasopharyngeal aspirate specimens from clinically suspected SARS patients collected during 1^st to 5^th day onset were tested by an automatic nucleic acid extraction system followed by a biotinylated single-tube nested polymerase chain reaction assay with microwell hybridization assay (bPCR-ELISA) for detection of SARS Coronavirus. A total of 628 specimens were collected for evaluation using three DNA amplification techniques: newly designed bPCR-ELISA, modified single-tube nested PCR (nPCR), and conventional two-step RT-PCR assay (cPCR).

Results: Using seroconversion as the gold standard for SARS diagnosis, the resolved performance of bPCR-ELISA, nPCR and cPCR was found at sensitivities of 79%, 61% and 68%, respectively. All PCR assays exhibited a 100% specificity. In evaluation of bPCR-ELISA, a clear distinction between PCR-positive and PCR-negative specimens when an OD405 value of 0.039 was chosen as cut-off. With the additional testing a second specimen and third specimen from same patient confirmed to have SARS CoV increased the overall sensitivity from 79% to 84 and 85%, respectively.

Conclusions: This study highlights the high throughput and performance of automatic RNA extraction in coordination with biotinylated single-tube nested PCR assay followed with ELISA products detection suitable for large-scale routine diagnosis in case of future SARS epidemic. Single technical personnel can process from extraction, amplification to detection and reported 96 samples within 24 hours. One additional sample tested may increase the sensitivity significantly.

Laboratory accreditation is an independent acknowledgement, by way of inspection, that a laboratory has the ability to provide quality, i.e. accurate, timely and consultative pathology diagnostic services. Accreditation schemes are designed to be educational and to lead to quality improvement but NOT to be a punitive activity. The inspectors use checklists to evaluate laboratory's compliance with the Standards. In Hong Kong, the accreditation programme for medical laboratories was launched by the Hong Kong Accreditation Scheme (HKAS) of the Innovation and Technology Commission, HKSAR Government earlier in 2004. The new ISO 15189 was adopted as the accreditation criteria. A number of medical laboratories in Hong Kong are holding current accreditation from overseas accreditation bodies such as the National Association of Testing Authority (NATA) of Australia or the College of American Pathologists (CAP) of the United States. These laboratories are accredited either to ISO/IEC 17025 or CAP standard. To encourage those medical laboratories to participate in the local scheme and to leverage their effort for attaining current accreditation, a bridging scheme was implemented in January 2005 by the HKAS for those laboratories holding overseas laboratory accreditation for a period of two years. The scheme aims to provide medical laboratories a user-friendly transition from compliance with ISO/IEC 17025 or other highly regarded standards in the medical field to ISO 15189. This is possible as ISO 15189 is based on ISO/IEC 17025 with additional elements relevant to medical laboratories such as laboratory safety, pre-examination procedures, post-examination procedures, determination of critical/alert intervals, turnaround times, etc. while for the CAP standards, some of the requirements are "missing" or being integrated into individual checklists a compare to those in ISO 15189. As these medical laboratories have already been assessed by other reputable accreditation bodies as being technically competent, HKAS could focus on the differences in specific requirements during the initial assessment. Only random checks on continual compliance with the common requirements are necessary. So, less effort is required for laboratories to update their system and documents. The duration of the assessment visit will correspondingly be shortened and the assessment fee will also be less compared with a full assessment. As all the elements in the requirements of ISO 15189 will be addressed, either via the previous assessment by other accreditation bodies or by HKAS Executive in the initial assessment, the HKAS standard for accreditation will be maintained.

Children acute lymphoblastic leukaemia (ALL) is a heterogeneous blood malignancy with diverse clinical outcomes. The stratification of patients to risk-directed therapy has contributed to substantial improvements in treatment. The TEL/AML1 fusion gene resulting from a subtle chromosomal translocation, t(12;21)(p13;q22), was recently characterized as a common genetic lesion and an independent prognostic factor in paediatric ALL. We applied nested reverse-transcription polymerase chain reaction (RT-PCR), cytogenetic analysis and flow cytometry to diagnostic bone marrow (BM) of 148 consecutive de novo childhood ALL at Prince of Wales Hospital from September 1995 to February 2005. We also monitored minimal residual disease (MRD) in the cohort of children with TEL/AML1 rearrangement. TEL/AML1 was detected in 16.5% (22/133) children with B-lineage ALL, of the median age of 4.5 years (range, 2 - 14) and white cell count of 8.5 x 10⁹/L (2.4 - 98.3 x 10⁹), but in none of 15 childhood T-lineage ALL. In addition, TEL/AML1 did not co-exist with E2a/PBX1, MLL/AF4 and BCR/abl fusion genes derived from t(1;19), t(4;11) and t(9;22), respectively. Immunophenotyping showed that TEL/AML1 was restricted to CD10⁺ B-precursor cell ALL. A higher incidence of aberrant expression of cross-lineage cellular antigens was noted in the TEL/AML1⁺ patient cohort although statistically insignificant (54.5% [12/22] of TEL/AML1⁺ ALL vs. 35.7% [45/126] of TEL/AML1^- ALL, p = 0.0094). Karyotyping demonstrated no discernible t(12;21)(p13;q22), nevertheless 36.4% (8/22) of TEL/AML1⁺ children showed either numeric or structural abnormalities. The modal number of six TEL/AML1⁺ children with abnormal cytogenetics ranged fro m43 to 48. Hyperdiploidy was found in 18.2% (4/22) children. The non=random ! t(12;21) (p13;q22) evident by TEL/AML1 was associated with trisomy 21, non-specific deletion of 6q and the loss of the normal TEL allele displaying the subtle 12p- in 4.5% (1/22), 18.2% (4/22) and 9.1% (2/22) children, respectively. At four weeks post-induction therapy rapid clearance of TEL/AML1 was noted in 68.2% (15/22) children. MRD was found in 8.3% (1/12) children with BM available at six months post-consolidation therapy. Upon completion of the maintenance therapy at 24 months, MRD was detected in 10.5% (2/19) children. None of the children with negative MRD post-induction therapy had TEL/AML1 detected later during the consolidation and maintenance therapy. To-date 18.2% (4/22) children relapsed at the median duration of 48.6 months (25.7-76.9) from diagnosis. Two of the four children with disease recurrence showed no MRD until relapse. The rates of overall event-free survival at the median time interval of 115 months of 22 TEL/AML1⁺ and 126 TEL/AML1^- ALL children were similar (94.4% vs. 84.1%, p = 0.206). The rates of disease-free survival were also comparable (70.6% vs. 76.8%, p = 0.599). Our data suggest that the rapid clearance of the fusion gene and the long duration of clinical remission in most of the children may attribute to the high sensitivity of TEL/AML1⁺ leukaemic clones to chemotherapy. However, less aggressive therapeutic regimen for TEL/AML1⁺ ALL was not recommended. RT-PCR for TEL/AML1 fusion genes remains to be useful and sensitive for detecting the subtle t(12;21)(p13;q22), which is hardly revealed by conventional cytogenetics, and monitoring MRD in this group of patients.

Cervical cancer is the fourth commonest cancer among women in Hong Kong and is the seventh leading cause of female cancer deaths. In fact, most cervical cancer can be prevented and stopped from progressing further with early medical treatment. As persistence of high-risk human papilloma virus (hrHPV) genotypes is a significant risk factor for the development of cervical cancer and found to be associated with more than 99% cervical cancer worldwide, detection of hrHPV genotypes will improve the sensitivity and specificity of routine cervical screening.

In this study, two licensed commercially available assays, Digene HC II test and Roche High-Risk HPV PCR test, for HPV detection were compared and evaluated as a screening tool for hrHPV detection in liquid based cytology (LBC) specimens. The PCR test is the first and the only PCR technology-based assay approved for in vitro diagnostic purposes. LBC specimens were taken for cytology, Digene HC II test and the Roche High-Risk HPV PCR test. The sensitivity and specificity of the two assays was compared. Positive and discrepancy results were confirmed by PCR-sequencing.

The currently available Digene Hybrid Capture II HPV test has been reported to have cross-reactivity with non-oncogenic genotypes and occasional indeterminate findings obtained. In this study, the hrHPV PCR test was able to pick up Digene HC II negative but hrHPV positive samples. Furthermore, the assay was able to identify specimens falsely reported by the Digene HC II test as high-risk HPV or low-risk HPV positive. In conclusion, the Roche hrHPV PCR test was more sensitive and specific in picking up hrHPV genotypes when compared with the Digene HC II assay.

Diagnostic molecular techniques have been gaining much popularity into clinical laboratories worldwide in recent years. With the promulgation of the new ISO 15189 standards for medical laboratory testing, the requirements of the quality management system, policies and procedures in relation to these new standards have thus brought about revised diagnostic molecular pathology practices in different medical laboratory disciplines.

It is the intention of this lecture to highlight those aspects of improved internal quality control measures and quality assurance issues pertaining to the practice of diagnostic molecular techniques in laboratory diagnosis. Specific focuses on nucleic acid extraction and purifications, method validation and optimization for Polymerase Chain Reaction (PCR) and Reverse Transcriptase - PCR (RT-PCR), safety issues, staff competency as well as external proficiency testing programs on diagnostic molecular pathology methods will be discussed. And whenever applicable, practical tips will be suggested.

Analysis of DNA sequence variations, either polymorphisms or disease-causing mutations, is crucial to biomedical science research and to molecular testing for the diagnosis of diseases. Denaturing high performance liquid chromatography (DHPLC) differentially separates DNA molecules, double- or single-stranded, on the basis of their interaction through amphilic ions with the non-polar stationary phase and the hydro-organic mobile phase. Under partially denaturing conditions, DHPLC can differentially separate homoduplexes and heteroduplexes carrying mismatches under suitable column temperature, which varies according to the sequence of the DNA fragments being analyzed. As such, DHPLC has now become a very powerful tool for screening and identifying unknown sequence variations. Its sensitivity and specificity of detecting sequence variations consistently exceed 95%. Under completely denaturing conditions, DHPLC separates short single-stranded DNA fragments according to both the length and base composition of the fragments. Thus, coupled with primer extension reaction, DHPLC allows rapid and reliable genotyping of known sequence variations. Multiplexing of 5-7 primer extension reactions can increase the throughput and is particularly useful in molecular diagnostics. On the other hand, DHPLC can also be used in conjunction with DNA pooling. Under partially denaturing conditions, appropriate DNA pooling can speed up the process of screening sequence variations. Under completely denaturing conditions, DNA pooling allows accurate estimation of relative allele frequencies of single nucleotide polymorphisms analyzed by primer extension coupled with DHPLC. This approach is extremely useful in mapping genes involved in complex diseases using genetic association studies. Development of monolithic capillary columns and parallel analysis can further increase the sensitivity and throughput of DHPLC.

Advances in medicine and new antibiotics have enabled the treatment of individuals who previously would have died, can now be treated and kept alive. After six decades of antibiotic use, the prevalence of antibiotic-resistant bacteria is increasing. Inappropriate antibiotic therapy has resulted in a greater incidence of resistant bacteria which may be difficult to treat. The best available ways to decrease and control antibiotic resistance are: rational use of antibiotics (e.g. appropriate selection of drug, dose, duration of treatment), good infection control procedures (hygienic practice and isolation), as well as local, national and global surveillance networks for monitoring dissemination of antimicrobial resistance and detection of new resistance mechanisms.

Microbiology laboratory has developed and made available cutting edge tools for the epidemiologic surveillance and outbreak investigation in support to the hospital infection control and antibiotic resistance control programmes. Research and development programmes have led to technological innovation in the fields of rapid bacterial identification, detection of resistance genes and epidemiological typing. Research in collaboration with clinical teams has focused on detection of emerging new pathogens and prevention of infections. Another major role is to participate in multicentre studies and epidemiologic surveillance networks.

Genital human papillomavirus (HPV) infection is causally associated with cervical cancer, a leading cause of cancer-related mortality in woman worldwide. Infection with high-risk type is the single most important risk factor for cervical intraepithelial neoplasia. Traditionally, the screening for cervical cancer is based on the manually prepared Pap smear. The sensitivity and specificity depend on the skill of the observer. In order to enhance the sensitivity for detecting cervical cancer, the fluid-based ThinPrep method (Cytyc Corporation, USA) and the nucleic acid based assay have been developed. The ThinPrep method can improve cytologic diagnosis of precancerous lesions compared with the conventional Pap method. The commercially available Hybrid Capture II (HC II) (Cytyc Corporation, USA) and polymerase chain reaction (PCR) methods using various primer pairs are validated and available for routine use. The PCR assay can distinguish the types of HPV whereas the HC II can differentiate between 13 high/intermediate risk and 5 low risk HPV types. The PCR methods are based on the amplification of HPV DNA. Two amplification methods were used. Firstly the type-specific DNA primers were used to detect the specific types of HPV. Secondly the amplification was done by different consensus primers system. The most common consensus primers systems are GP5+/6+ PCR, SPF1/2 PCR and MY09/11 PCR. The SPF1/2 system has the highest analytical sensitivity among these systems.

The present study evaluates the usefulness of HPV PCR and HC II as a screening tool from ThinPrep cytology specimens. Both assays were performed in a clinical study comprising 40 women. Results obtained by the HC II assay were compared with a PCR amplification strategy, which included type specific and consensus PCR primer systems (GP5+/6+ PCR and SPF1/2 PCR). Results indicating that HPV DNA was detectable among the cohort of 40 women by the PCR and HC II in 57.5% (23/40) and 40.0% (16/40) respectively. The overall concordance was 77.5%. The hybrid capture II (HC-II) sensitivity and specificity versus PCR was 65.2% and 94.1%. PCR sensitivity and specificity versus HC-II were 93.8% and 66.7%. For sample with normal cytology, the agreement between both assays was 70.8%. Sensitivity, specificity, positive predictive value and negative predictive value of the HC-II test were 41.7%, 100.0%, 100.0% and 63.2% respectively. For samples with abnormal cytology, the level of agreement was 87.5%. Sensitivity, specificity, positive predictive value and negative predictive value of the HC-II test were 90.9%, 80.0%, 90.9% and 80.0% respectively. The results of PCR and the HC-II assay indicated good reproducibility. For sample with abnormal cytology, the concordance was good and resulting in substantial reproducibility.