Engaging Multiple Stakeholders To Reduce Adverse Drug Reactions
Health Sciences Authority

A. Problem Analysis

 1. What was the problem before the implementation of the initiative?
The Vigilance and Compliance Branch (VCB) of the Health Sciences Authority (HSA) is the government agency responsible for collecting adverse drug reaction (ADR) reports from across the healthcare institutions in Singapore and mitigating the risks associated with specific drugs. Over the years, HSA has observed that cutaneous/skin reactions comprise approximately half of all reported ADR cases. These reactions span from relatively mild reactions such as rash to very serious and unpredictable reactions such as Stevens-Johnson syndrome (SJS) and Toxic Epidermal Necrolysis (TEN), which are ADRs associated with high morbidity and up to 30% mortality. Moreover, almost all patients who suffer SJS or TEN, have at least one sequelae that affects daily quality of life for many years after the acute episode. Sequelae involve skin, eyes and mucous membranes (such as oral, gastrointestinal, pulmonary, genital as well as urinary) and require interdisciplinary care. Pharmacogenetics is the study of variations in DNA sequence as related to drug response, either efficacy or toxicity. With advances in genomic technologies, a growing number of genetic variants have been linked to an increased risk of an ADR. However, the variants linked to ADRs are often different in different ethnic populations. Although HSA was interested to apply pharmacogenetic knowledge to improving the safe use of medicines, there was no infrastructure in place to readily assess or confirm such findings in the ethnic group in Singapore because ADR cases reported to HSA have limited clinical data and no biological samples. In response to this situation, HSA took the initiative to lead and coordinate with the hospitals to collect detailed clinical data and blood samples of serious skin rash ADR cases. A national effort was necessary to accrue enough patient samples within a reasonable time frame to assess these findings scientifically. It was an extensive effort to bring many parties from various institutions and disciplines together (including researchers, clinicians, and hospitals) towards a common goal to advance public health and to reduce the unpredictability of selected serious ADRs for improved patient safety. The objective was to investigate if there was a genetic marker valid for the ethnic groups in Singapore that could identify patients susceptible to certain drug-induced serious skin reactions. This genetic marker could then be used to prospectively identify individuals at risk of a severe drug-induced skin reaction so that they could be given an alternative drug, thereby reducing the risks of ADR associated with that drug and improving medication safety.

B. Strategic Approach

 2. What was the solution?
Chairman of the HSA Board at the time, Prof. Edison Liu, proposed that HSA consider applying genomic knowledge towards optimizing drug therapy. HSA established an advisory expert panel and a working group to evaluate different options to operationalize this concept. After extensively studying the literature in this field and engaging in discussions with local scientists and clinicians and international key opinion leaders, HSA decided to pilot a program to establish a case-control collection of blood samples from patients who had experienced serious ADRs from marketed drugs. By gradually building up a repository of biological samples and clinical data from such cases, HSA would position itself to validate findings reported in other ethnic populations and possibly even to discover new genetic associations. Based on ADR records received by HSA, at least 15% of all serious skin reaction reports were attributable to carbamazepine (CBZ). Notably, 25% of CBZ cases affected Malays, though they comprise only 13.6% of the general population. In 2004, Taiwan published a landmark discovery of a highly significant genetic association between HLA-B* 1502 and CBZ-induced SJS and TEN among Han Chinese. However, the allele was not a predictive biomarker in Caucasian or Japanese populations. Due to variability in the association for different ethnic populations and even within different Asian groups, HSA recognized the need for data on Singaporean ethnic groups to inform a rational national pharmacogenomics policy. Hence, HSA embarked on a program to develop infrastructure for collection, storage, and analysis of DNA samples from patients who had experienced serious skin reactions as well as from drug-tolerant controls and to capture demographic and clinical data associated with those samples. The program has now been set up at four public hospitals: the National University Hospital (NUH), Singapore General Hospital (SGH), Changi General Hospital (CGH) and the National Skin Center (NSC). Between March 2009 and November 2012, 158 cases of drug-induced SJS/TEN were screened and 47 cases enrolled and completed the study. Altogether, 13 CBZ cases and 26 CBZ-tolerant controls were collected and genotyped. All 13 CBZ cases, but only 3 controls, tested positive for HLA-B*1502 for a highly significant odds ratio of 181. Genotyping of 609 healthy individuals in the general population, performed by research institutes and the National University of Singapore, revealed that the allele is present in approximately 1 in 5 Malays, 1 in 8 Chinese, and 1 in 25 Indians in Singapore. In November 2012, HSA convened an expert panel meeting to review the data and discuss the impact on clinical practice. The panel concluded that based on the strong local and international data supporting an association between HLA-B*1502 and CBZ-induced SJS/TEN, testing of HLA-B*1502 for new CBZ users should be recommended. However, outstanding issues such as turnaround time and the cost of testing were identified as barriers to adoption. The Molecular Diagnosis Centre (MDC) at NUH established a centralized testing laboratory for the HLA-B*1502 genotype, hence lowering the cost of the test and reducing the turnaround time. Given the strength of the scientific evidence and cost-effectiveness of testing (see Q4), the Ministry of Health (MOH) offered to provide a 75% subsidy for the genotyping test for subsidized patients at public clinics and hospitals. After another consultation session with stakeholders, a Dear Health Care Professional Letter was disseminated to physicians in April 2013 informing them that genotyping for the HLA-B*1502 allele prior to the initiation of CBZ therapy in new patients of Asian ancestry would be considered the new standard of care. Since then (21 months), HSA has not received any reports of SJS/TEN associated with the use of CBZ in patients screened for the HLA-B*1502 allele.

 3. How did the initiative solve the problem and improve people’s lives?
Prior to the HSA initiative, it was difficult to undertake studies of pharmacogenomics associations of ADRs locally because there were insufficient numbers of cases at any one healthcare institution. Without a national-level coordinated effort to involve multiple hospitals, it would not be possible to collect adequate numbers of drug-induced SJS/TEN in a reasonable timeframe. HSA successfully coordinated a multi-institutional collection, collaborated with scientists to analyze the samples and evaluate cost-effectiveness, and held consultation sessions with relevant stakeholders to elicit input and address implementation issues. This initiative made it possible to formulate a nationally relevant genotyping recommendation, reduce the cost and turnaround time of genotyping, and obtain subsidies for lower-income patients. The implementation of HLA-B*1502 genotyping prior to CBZ therapy in Singapore provides a model for how a government agency can promote collaboration in the scientific and clinical community towards a common goal to reduce the unpredictability of serious ADRs and contribute towards safer use of medicines .

C. Execution and Implementation

 4. In which ways is the initiative creative and innovative?
*Staff Training At the time of initiation of the project, research and pharmacogenetics (PGx) were relatively new to HSA, and there was a need to build the PGx team’s knowledge and capability in these areas. HSA invited key experts from academic, research and clinical sectors to form a Pharmacogenetics Expert Panel (PEP) to provide guidance on setting up the infrastructure in the area of PGx. Foreign experts who had conducted similar PGx studies were consulted for their input. The PGx team also organized a 12-week course “Principles of PGx for Drug Regulators” consisting of weekly lectures by PEP members and other local experts on various topics related to PGx to develop HSA staff knowledge in this emerging field. *Case-Control Collection HSA initiated collaborations with four public hospitals to recruit and collect DNA samples from patients who had experienced SJS/TEN. Samples were also obtained from drug-tolerant controls, patients who had taken the same drug but had not experienced an ADR. HSA also established a partnership with biobanking facilities that could extract and store coded DNA and plasma samples. High-resolution genotyping was performed through collaboration with the Singapore Immunology Network (SIgN). The results confirmed that carriers of HLA-B*1502 had a 181-fold higher risk of developing SJS/TEN when given CBZ compared with non-carriers. *Cost-effectiveness study Concurrent with the case-control collection, HSA undertook a study in collaboration with the Duke-NUS Health Services and Systems Research Program to evaluate the cost-effectiveness of genotyping new epilepsy patients prior to selecting an antiepileptic drug. The analysis concluded that genotyping new epilepsy patients in Singapore for HLA-B*1502 is more cost effective than prescribing or avoiding usage of CBZ without knowledge of the genotype. *Extensive stakeholder consultations Consultations with physicians highlighted that cost and turnaround time were barriers to adoption of genotyping. These were resolved through the establishment of a centralized HLA-B*1502 genotyping facility. HCPs were informed on 30 April 2013 that “genotyping for HLA-B*1502 allele prior to initiation of CBZ therapy in new patients of Asian ancestry is now considered the standard of care” and that a 75% subsidy for the genotyping test was available for subsidized patients at public clinics and hospitals. As of 30 September 2014, a total of 1057 blood samples have been tested for HLA-B*1502 genotype, and 110 (10.4%) samples were positive for the presence of the HLA-B*1502 allele. It is noteworthy that HSA received 131 local serious reports of CBZ-induced SJS/TEN (an average of 15 reports of CBZ-induced SJS/TEN per year) between 2003 to 2012. In contrast, HSA has not received any reports of SJS/TEN associated with the use of CBZ in patients screened for the HLA-B*1502 allele in the 21 months since issuing the new guidelines on HLA-B*1502 genotype screening. One of the advantages of testing is that patients who are not carriers of HLA-B*1502 have a very low risk of SJS/TEN. Now physicians have greater confidence prescribing CBZ, a very inexpensive anti-epileptic drug, to test-negative patients. This is an economic benefit to epilepsy patients who may have to take medication for the rest of their lifetime.

 5. Who implemented the initiative and what is the size of the population affected by this initiative?
The HSA PGx Team consists of staff of the Vigilance and Compliance Branch (VCB) under the Health Products Regulation Group (HPRG). Chan Cheng Leng (CCL) is Group Director HPRG; Dorothy SL Toh (DSLT) is Director 1 VCB. The HSA PGx Team additionally consists of Cynthia Sung, PGx Team Leader (CS), Liesbet L. Tan (LLT), Celine Loke (CL), Michael Limenta (ML) and Tan-Koi Wei Chuen (TKWC). Study design was conceived by CS, DSLT, CCL, LLT; Clinical sites were set up by CS, DSLT, LLT, CL; Monitoring of clinical sites was performed by LLT and CL. The entire PGx team was involved in consultation sessions with scientific and clinical community. Clinical collaborators who recruited cases or controls are from four public hospitals: ➢ NUH: Derrick C.W. Aw, Aisha Lateef, Chan Yee Cheun, June Tan,Teoh Hock Luen ➢ SGH: Pang Shiu Ming, Lim Shih Hui, T. Thirumoorthy, Lee Haur Yueh, Koh Hong Yi, Lui Nai Lee, Wong Siaw Ing, Chong Yong Yeow and Chan Choong Meng ➢ CGH: Tay Yong Kwang, Tan Siew Kiang, Archana Vasudevan, Lee Shan Xian ➢ NSC: Lim Yen Loo, Heng Yee Kiat Clinical research coordinators assisting with patient recruitment and data collection: ➢ NUH: Seow Lih Jen, Serene Ng, Eric Seow ➢ SGH: Goh Poh Kuan, Ei Mon Soe ➢ CGH: Low Li Fang, Geraldine Lim, Candy Yang ➢ NSC: Jolene Ong, Kwa Jie Si STN staff assisting with development of electronic case report forms: Lee Yan Lim, Tzer Jing Seng Other collaborators include ➢ Edmund JD Lee (Department of Pharmacology, NUH) chaired the HSA PEP that advised the PGx Team on this initiative. ➢ Ren Ee Chee (Singapore Immunology Network) performed the high resolution HLA typing of the samples. ➢ Evelyn S.C. Koay (MDC, NUH) established the centralized laboratory of a validated genotyping test for HLA-B*1502.
 6. How was the strategy implemented and what resources were mobilized?
This initiative was funded by the HSA Reinvestment Fund, which was used to pay for part of the salaries of clinical research coordinators at the public hospitals, for transport of samples to the biobanking facility and for DNA and plasma extraction and storage. Since inception, HSA has spent approximately S$51,000 to date on this initiative, in addition to 1.5 FTEs per year of HSA staff time. The cost-effectiveness analysis was performed for free by a graduate student at Duke-NUS for a thesis project. Genotyping costs for the case-control samples were contributed in kind by the Singapore Immunology Network. Validation costs for the clinical genotyping test were absorbed by the Molecular Diagnosis Center.

 7. Who were the stakeholders involved in the design of the initiative and in its implementation?
This initiative successfully created a network of clinicians at public hospitals and researchers that HSA could rely upon to assist in studying ADR cases and applying a new technology to reduce ADR incidence. The clinicians had registries of patients who had suffered such serious skin reactions, and the nurses took the time and effort to trace and contact these patients to ask if they would like to participate in the study. In addition, the clinicians were dedicated in recruiting new SJS/TEN patients from their hospitals to take part in this study. HSA undertook a study in collaboration with the Duke-NUS Health Services and Systems Research Program to evaluate the cost-effectiveness of genotyping new epilepsy patients prior to selecting an antiepileptic drug. The study found that genotyping new epilepsy patients in Singapore for HLA-B*1502 is more cost effective than prescribing or avoiding usage of CBZ without knowledge of the genotype. This report played an important role in deliberations by MOH and other stakeholders as they formulated new practice guidelines and approved a subsidy for low income patients. Translation of PGx into patient care is challenging because it involves changing established medical practice. In this instance, the evidence that the genetic marker HLA-B*1502 was very strong based not only on international data, but local data as well. Nonetheless, there were issues of costs and availability of a testing facility to offer this service to screen patients for the HLA-B*1502 marker. The establishment of a single centralized laboratory offering the test was considered the most effective way to reduce both the turnaround time and genotyping cost. In addition, the availability of the subsidy was instrumental in the success of this initiative. All these factors contributed towards smooth adoption of the new standard of care – that genotyping for the HLA-B*1502 allele prior to the initiation of carbamazepine therapy in new patients of Asian ancestry would be the new standard of care. With the new guidelines, patients at high risk of SJS/TEN with CBZ are offered alternative anti-epileptic drugs, while test-negative patients can be prescribed a low cost, effective medicine with very low risk of developing SJS/TEN. The most gratifying output of this programme is the elimination of cases of CBZ-associated SJS/TEN when genetic testing is done.

 8. What were the most successful outputs and why was the initiative effective?
Regular audits and visits were conducted at the different centers to ensure accuracy of the data collected and to address operational issues. Letters were sent to the clinicians after every visit to keep them informed on the progress of the study and whether there were any issues with the conduct of the study. Periodic reminders were sent to the clinical research coordinators to stay alert for cases of SJS/TEN so that the program would steadily accrue cases for the study. The HSA PGx team has convened 6 meetings of the Pharmacogenomics Expert Panel (PEP) and met more frequently with the PEP Chairman to update our advisors on the progress of the program and receive their input on implementation issues and advice about other stakeholders HSA should involve or consult. Progress reports on the PGx initiative were included in annual reporting to senior management.

 9. What were the main obstacles encountered and how were they overcome?
The PGx team established a partnership with the Singapore Tissue Network (STN) as the centralized repository for patient samples and clinical data. STN was selected because it operated independently of any one hospital and had achieved ISO9000 quality certification. However, in 2011 the parent organization of STN (by then renamed the Singapore Biobank) decided to close the facility. HSA had to review the various options for biobanking facilities in Singapore, and then transfer all the samples and data to the newly selected facility, the NUH Tissue Repository. When it became evident that the genetic association between HLA-B*1502 and SJS/TEN among new CBZ users was very strong for the local population, and that it was cost effective to perform genotyping before prescribing or avoiding usage of the drug from an overall healthcare system perspective, HSA sought input from clinicians on how to implement genetic testing into clinical practice. From the consultations, HSA learned that the cost of the test (at the time S$350) would be a significant obstacle to patients, especially low-income patients. Through the discussions, consensus was reached that having a centralized laboratory would reduce costs through economies of scale. The Molecular Diagnostic Centre (MDC) of NUH took up this role and was able to reduce the genotyping test to S$200 and reduce turnaround time to 2 to 4 days, down from 7 days. At this juncture, the Ministry of Health lent its support by providing a 75% for subsidized patients at public clinics and hospitals. The public consultation session allowed HSA to understand barriers to adoption of a new practice guideline and to proactively find a solution to the problems. This ensured better acceptance and smoother implementation of the new clinical practice.

D. Impact and Sustainability

 10. What were the key benefits resulting from this initiative?
Prior to issuance of the Dear Health Care Professional Letter (DHCPL) in April 2013, HSA received an average of 15 cases of CBZ-associated SJS/TEN per year. In the 21 months since then, HSA has not received any reports of SJS/TEN associated with the use of CBZ in patients screened for the HLA-B*1502 allele. A total of 1057samples have been sent for testing as of 30 September 2014, and 110 (10.4%) of the samples were positive for HLA-B*1502. It appears that a one-time HLA-B*1502 genotyping test restores confidence among clinicians in using CBZ as it helps differentiate high-risk patients who should avoid CBZ from low-risk patients who are able to take this low-cost yet effective medicine. Availability of an inexpensive drug like CBZ for life-long treatment of epilepsy translates into substantial savings for the patients. The benefits of personalized medicine are exemplified by the new practice. Another important benefit of this initiative is that HSA has built a network of clinicians and scientists who are willing to work together towards a common goal of improving patient safety. HSA has demonstrated that it is capable of coordinating a complex multi-institutional, multi-disciplinary program for public health objectives. HSA’s experience has stimulated interest from other institutions to conduct other PGx programs on ADRs. Scientists from those institutions are now assuming a larger role in performing case-control studies. The Singapore Agency for Science and Technology has approved funding for a multi-institutional study “The Surveillance And Pharmacogenomics Initiative for Adverse Drug Reactions (SAPhIRE): a strategic partnership involving HSA, the Biomedical Research Council, and Hospitals to improve drug safety.” BMRC-funded research institutes such as The Genome Institute of Singapore and the Translational Laboratory for Genetic Medicine, in collaboration with the hospitals, are taking the lead on collections of other ADRs such as statin-induced myopathy and chemotherapeutic drug-induced hearing loss and cardiomyopathy. HSA’s decision on genotyping was supported with objective scientific data from international and local studies. It listened to the concerns of the clinicians and worked proactively to resolve issues they raised. At the first consultation session that HSA held with clinicians, cost and turnaround time of the genotyping test were identified as significant barriers to routine genotyping. The only laboratory offering the test in the nation was HSA’s tissue typing laboratory, which charged S$350 and required 7-days to return results. To avoid the perception of a conflict of interest, HSA felt it was important to avoid advocating routine testing of a test that could only be performed by an HSA laboratory. Ultimately, the Molecular Diagnostics Center at the National University Hospital was able to offer the test at S$200 per test and achieve a turnaround of 2-4 working days. MOH approved a 75% subsidy for low-income patients, which helped cushion the cost to low-income patients.

 11. Did the initiative improve integrity and/or accountability in public service? (If applicable)
As noted in Question 10, other organizations in Singapore are now devoting resources to conducting similar case-control pharmacogenetic studies. The ADRs they plan to study are statin-induced myopathy and chemotherapy drug-induced hearing loss and cardiomyopathy. These organizations have learned from HSA’s leadership and example that multi-institutional collaborations and stakeholder engagement are the key to success of such programs. HSA’s collection of serious skin ADRs cases continues to accrue cases. The program has now expanded recruitment to include other cutaneous ADRs, namely drug rash with eosinophilia and systemic symptoms (DRESS) and acute generalized exanthematous pustulosis (AGEP). HSA has sufficient number of cases of allopurinol SJS/TEN/DRESS and drug-tolerant controls, which now have been genotyped, and a pharmacoeconomic analysis for genotyping is nearly completed. The data on Singapore’s population can also be used in collaboration with other international groups who have similar ADR collections to determine if the genetic associations are valid across a broad spectrum of populations or applicable only for certain ethnic groups. The PGx initiative to collect local ADR cases and drug-tolerant controls and the related cost-effectiveness analysis of prospective testing is being used as a model for public health policy formulation and stakeholder engagement for future cases of PGx associations with drug safety, to establish relevance in the local population and translation into clinical practice.

 12. Were special measures put in place to ensure that the initiative benefits women and girls and improves the situation of the poorest and most vulnerable? (If applicable)
From a scientific and health resource perspective, we learned that certain genetic markers need to be validated for specific populations. For example, the HLA-B*1502 marker is a predictive marker for Singapore’s ethnic groups, but not for Japanese or Europeans. Therefore, data for relevant ethnic groups should be known to support local public policy decisions. We also learned that genetic technologies can improve the benefit –risk profile of a drug, and that the clinical and economic merits must be evaluated on a case-by-case basis. From an implementation perspective, HSA learned about the various operational issues in the conduct of a PGx study to investigate the genetic predisposition for serious side effects to certain drugs in the major ethnic groups in Singapore. We learned that a public agency can play a significant role in bringing together professionals of diverse knowledge sets and coordinating a multi-disciplinary effort to accomplish a goal that advances public health and safety. By clearly articulating a public health goal, we found that professionals will step forward to assist even when institutional boundaries do not always make it easy. Good communication, transparent processes and follow-through fostered trust and cooperation from the various stakeholders in this project. These elements are essential for the success of future programs.

Contact Information

Institution Name:   Health Sciences Authority
Institution Type:   Government Agency  
Contact Person:   Cheng Leng CHAN
Title:   Health Products Regulation Group  
Telephone/ Fax:   +65 6866-3528/6467-9069
Institution's / Project's Website:  
E-mail:   chan_cheng_leng@hsa.gov.sg  
Address:   11 Biopolis Way, #11-03 Helios
Postal Code:   138667
City:   Singapore
State/Province:   Singapore

          Go Back

Print friendly Page