Implementation of Pre-emptive Pharmacogenomics Testing in Outpatient Clinics in Asia (IMPT study)

Introduction

Pharmacogenomics (PGx) is the study of how an individual’s genetic makeup influences their response to medications. One approach, which is preemptive PGx testing using multi-gene panels, has demonstrated clinical benefits in several studies, primarily conducted in the United States (US) and Europe (Dunnenberger et al., 2015; Gottesman et al., 2013; Ji et al., 2016; Swen et al., 2023; Tsermpini et al., 2020; Wang et al., 2022). In a study conducted among seven European countries encompassing both primary and tertiary care settings, over 90% genotyped patients had at least one actionable PGx variant, and preemptive PGx reduced clinically relevant adverse drug reactions by 30% (Swen et al., 2023). Clearly, preemptive PGx testing and implementation have major impacts on health outcomes that would warrant further evaluation and optimization.

With over 60% of patients visiting primary care in both the UK and the US receiving medications associated with PGx recommendations (Kimpton et al., 2019; Rollinson et al., 2020; Schildcrout et al., 2012), it is vital to fine tune the implementation of preemptive PGx in primary care. One study conducted in the Netherlands revealed that 1 in 19 new prescriptions in the primary care settings would require a dose adjustment or a change in drug therapy (Bank et al., 2019), underscoring the potential utility of preemptive PGx testing in delivering more efficient and effective patient care.

The implementation of PGx into the primary care setting is a multifaceted process that encompasses preemptive PGx research, the integration of PGx into clinical workflows, PGx testing, assessing clinical impact and providing PGx education (Giri et al., 2019). In view of the limited data related to preemptive PGx testing in the outpatient setting, we designed a study to assess the feasibility of implementing preemptive PGx services at outpatient clinics, encompassing both primary and secondary care settings in Singapore. We aimed to explore the practicality and potential challenges of integrating PGx testing, with a particular focus on understanding how the implementation of PGx might impact clinical workflows and patient care.

Methods

Intervention

In this study, we aimed to implement preemptive PGx at outpatient clinics within RMG. Our implementation of PGx testing involved a comprehensive strategy to ensure its successful integration into clinical practice. Our intervention encompassed training, patient recruitment, sample processing, and post-test reporting (Figure 1).

• Download figure
• Open in new tab

Figure 1:

Integrated Workflow: Training, Patient Recruitment, Sample Processing, and Reporting

a) Training

All participating clinicians (physicians and pharmacists) were required to attend a Pharmacogenomic workshop conducted by Nalagenetics’ Medical Affairs team, which is a comprehensive training program which covers various aspects related to the implementation of preemptive PGX, which includes the significance of PGx testing, the testing process, and PGx results interpretation (Adesta et al., 2021). Following the workshop, participating clinicians were expected to complete a post-training assessment to evaluate their use of PGx in clinical practice. Additionally, all clinic staff underwent specimen management training conducted by Nalagenetics’ Operations team, which included training on sample collection and operational workflow.

b) Patient Recruitment

The eligibility criteria included patients aged 21 to 65 with a reported history or risk of developing any of the target chronic conditions: diabetes mellitus, hypertension, dyslipidemia, ischemic heart disease, stroke, osteoarthritis, rheumatoid arthritis, gout, anxiety and major depressive disorder, or any patients who were prescribed with one of the 29 PGx-associated medications (Table 1). Patients unable to provide informed consent and/or those not meeting all eligibility criteria were excluded from the study. Patients eligible for the study underwent pre-test counseling conducted by eight physicians, followed by the administration of informed consent. Buccal samples were collected using OraCollect•DNA (DNA Genotek) under the supervision of trained clinic staff.

View this table:

• View inline
• View popup

Table 1: List of drugs, genes, and variants included in the pharmacogenomics panel.

c) Sample Processing

All collected samples were sent to the central laboratory, Raffles Diagnostica Lab, located at Raffles Hospital. Courier service was arranged twice a week for sample delivery to NalaGenetics’ Laboratory for testing.

We utilized the Nala RxReady™ panel to process the buccal samples (∼1mL) in batches, with each batch containing three patient samples. Nala RxReady™ is a multi-gene qPCR-based panel of 21 allele variants (Table 1), comprising five pharmacogenes - CYP2C9, CYP2C19, CYP2D6, SLCO1B1, and HLA-B*58:01. Genomic DNA was extracted using the Monarch® Genomic DNA Purification Kit. Subsequently, all DNA samples were analyzed using the Bio-Rad CFX96 IVD Touch™ Real-Time PCR Detection System, following the manufacturer’s instructions. The generated files were then imported into the companion software, Nala Clinical Decision Support™ (Nala CDS™), for in-depth analysis encompassing variant genotyping, diplotype determination, and phenotype translation. The clinical recommendations generated by the software were derived from comprehensive annotations available in Clinical Pharmacogenetics Implementation Consortium (CPIC), Dutch Pharmacogenomics Working Group (DPWG), Canadian Pharmacogenomics Network for Drug Safety (CPNDS), and regulatory bodies including the US Food and Drug Administration (FDA), Pharmaceuticals and Medical Devices Agency (PMDA), Swissmedic, and European Medicines Agency (EMA) (Kothary et al., 2021).

d) Post-test Report

Given the high level of integration of clinical workflows with the EMR system, PGx results including genotypes, phenotypes, and follow-up recommendations were seamlessly integrated into the EMR system to facilitate the study. As depicted in Figure 2 below, a summary of patients’ PGx results automatically generates when their PGx results became available for the first time. This summary can also be accessed through the “Patient Alert” tab. Drugs with follow-up recommendations were prioritized and prominently displayed on the page. The full report (Supplementary Material S1), containing clinical recommendations, can be retrieved by selecting the recommendation hyperlink to guide prescribing.

• Download figure
• Open in new tab

Figure 2:

Screenshot of the EMR interface. The summary comprises the drug names, gene list, phenotypes, follow-up recommendation symbols and a hyperlink to access the full report.

All participating physicians were also granted access to the Nala CDS™ Platform, enabling them to view and release PGx reports for patients (Figure 3). Patients can then access their PGx results through the Nalagenetics’ Mobile application (Figure 4). Post-test counseling was offered by the physician who ordered the test during the next follow-up visits or in response to patients’ requests.

• Download figure
• Open in new tab

Figure 3:

User interface on Nala CDS Platform.

• Download figure
• Open in new tab

Figure 4:

Mobile app interface for PGx report access.

Results

Patient Recruitment and Demographics

Out of the 1428 eligible patients, 224 patients (15.7%) were recruited by 8 clinicians over an 11-months period. One patient dropped out, and one was lost to follow-up during resampling, resulting in an analyze sample of 222 patients (Figure 5). The study included 222 patients, comprising 126 females and 96 males, with an average (SD) age of 43 (10.9) years. Majority of the participants were Chinese (61.3%) by racial/ethnic origin, followed by other Asians (18.9%), Indian (9.5%), Malay (5.9%), and non-Asians (4.5%) (Table 2). Slightly over half of the patients (50.9%) self-reported to be using at least one medication.

• Download figure
• Open in new tab

Figure 5:

Flowchart of study participants and recruiting physicians.

View this table:

• View inline
• View popup

Table 2: Demographics of recruited participants.

Discussion

This is one of the first studies that explores the feasibility of implementing preemptive PGx testing in an Asian primary care setting. Almost all (95%) analyzed patients carried at least one clinically actionable variant using our multigene panel, suggesting the potential clinical benefits of preemptive PGx testing. This finding aligns with other studies (Smith et al., 2022; Swen et al., 2023; Van Driest et al., 2014) that were conducted in the US and in Europe. Our study also reveals that over half of the patients have taken or are currently taking at least one PGx-associated medication, with one out of 5 receiving follow-up recommendations, including changes in prescription, dosage adjustments, and increased monitoring. Our findings align with the findings of another study conducted in the Netherlands (Bank et al., 2019), where one out of 5 new prescriptions within the primary care setting were associated with actionable PGx recommendations. Collectively, these findings emphasize the potential of PGx testing to optimize patient care and underscore its relevance in personalized medicine.

We have also observed significant variations in drug metabolizing phenotypes among different ethnic groups within our study population. It is observed that over 30% of the study population, excluding Indians (17%), may exhibit the CYP2D6 Intermediate Metabolizer phenotype, thereby warranting actionable recommendations for medications such as antidepressants, tamoxifen, and flecainide. Given the high prevalence of CYP2C19 intermediate metabolizer phenotypes among Asians in Singapore, particularly within Indians, over 70% of the Indian population may necessitate an alternative antiplatelet therapy to clopidogrel. Furthermore, approximately 50% of non-Asians may exhibit the CYP2C19 Ultrarapid Metabolizer phenotype, hence requiring dosage adjustment or alternative therapies for specific antidepressants, proton pump inhibitors, and voriconazole. Asians, excluding Indians, also demonstrate a higher likelihood of developing side effects to Simvastatin (SLCO1B1) and Allopurinol (HLA-B* 58:01) compared to non-Asians (Whirl-Carrillo et al., 2021).

We have also observed a low psychological impact to patients on the return of PGx testing, as evidenced by the majority of participants feeling relieved and happy upon receiving their results. Such observation is similar to findings from other studies (Haga et al., 2014; Jameson et al., 2021) that PGx testing reports do not cause major psychological impact. It is important to acknowledge that PGx testing results can present challenges in interpretation due to their inherent complexity, particularly for patients with limited knowledge of genetics. This may prompt patients to seek further information despite having received post-testing counseling. Nonetheless, for PGx testing to effectively benefit both the patient and future healthcare providers, it is crucial for patients to fully comprehend their test results and the test’s purpose through pre- and post-test counseling. This ensures that patients can confidently share their PGx results with other healthcare providers, thereby facilitating personalized care (Haga et al., 2014).

It was interesting to observe a differing perspective between clinicians and patients regarding the cost of PGx testing, with patients having fewer concern about paying out-of-pocket costs for PGx testing. This finding contrasts with the prevailing notion that cost has consistently acted as a perceived barrier to PGx testing (Chan et al., 2017; Frigon et al., 2019). In this context, it is plausible that patients tend to prioritize the direct benefits of PGx testing, while clinicians consider factors such as clinical evidence, cost-effectiveness, and the potential financial strain on their patients. Given that the willingness to bear out-of-pocket costs for PGx testing is significantly influenced by individuals’ financial circumstances (Bielinski et al., 2017), it is noteworthy that this study was conducted in a private healthcare setting, where patients visiting private healthcare settings in Singapore might have a higher socioeconomic background.

Implementation of PGx testing in clinical practice is a complex process which encompasses PGx testing, EMR integration, and clinical impact of PGx (Giri et al., 2019). This study was also designed to explore and address some of these key considerations as part of the implementation process. To date, the sole mandatory PGx test in Singapore is genotyping for the HLA-B*15:02 allele before initiating carbamazepine (HSA, 2013). Limited studies described their experience with implementing PGx testing in healthcare settings in Singapore, with one describing the feasibility to implement preemptive PGx implementation in a primary care setting with all genotyped patients have at least one actionable PGx variants (Smith et al., 2022). In our study, the multigene panel was curated specifically to include alleles that were pertinent to Asian populations with a minor allele frequency >1% based on East Asians, South Asians, and Europeans obtained from the PharmGKB database (Kothary et al., 2021). The expected TAT of PGx testing was set at 5 business days because a long TAT may potentially lead to treatment delays. In this study, we have observed that >90% successfully met the targeted TAT. While meeting the expected TAT for PGx testing is crucial, conducting PGx testing preemptively may serve as a solution by identifying genetic factors that influence drug responses in advance. EMR integration plays a pivotal role in the successful implementation of PGx into healthcare systems (Giri et al., 2019), with majority of our clinicians (67%) preferred the usage of an EMR-integrated module over a standalone CDS platform. By seamlessly incorporating PGx results into EMR systems, this facilitates quick retrieval and display of relevant PGx data where healthcare professionals can access and utilize genetic information to make more informed and personalized decisions on patients’ medication management. Furthermore, PGx data becomes readily available for future clinical encounters, further enhancing the quality of patient care. Empowering patients in their healthcare decisions is also a crucial aspect of personalized medicine, and providing a dedicated mobile app would allow patients to gain direct access to their PGx results, thus allowing them to participate in their health management. Patients can easily review their PGx data, understand how it impacts their medication responses, and explore recommendations for personalized treatment. This encourages patients to be more proactive in discussions with their healthcare providers and to ask informed questions about their prescriptions, thus promoting shared decision-making.

There are major clinical implications with the findings of our study. We experience a low recruitment rate, with only a fraction of eligible patients enrolling in the study. Despite the potential benefits of PGx testing in optimizing medication therapy and improving patient outcomes, the challenges associated with recruitment highlight barriers that need to be addressed. For example. many patients were unfamiliar with PGx testing and its potential to guide medication selection and dosing. Therefore, efforts to increase awareness and education about the benefits of PGx testing are essential for successful implementation of PGx testing in the outpatient setting (Kabbani et al., 2023). Additionally, the observed emphasis on the need for clearer guidelines and insurance coverage to promote wider adoption of PGx testing in primary and secondary care aligns with the expected priorities in this field (Chan et al., 2017). While two major consortia, CPIC and DPWG provide valuable guidance, disparities in clinical recommendations arise from their distinct methodologies (Krebs & Milani, 2019). Furthermore, the absence of standardization in reporting formats presents a challenge (Lanting et al., 2020). Hence, a concerted effort is imperative to effectively address these issues. Furthermore, despite the growing interest in and the decreasing cost of PGx testing, cost remains a significant barrier to its implementation (Chan et al., 2017; Frigon et al., 2019). Therefore, expanding insurance coverage for PGx testing may serve as a key solution to address this barrier effectively. Previous studies have shown that patients are more inclined to undergo PGx testing when the costs are fully or partially covered by insurance (Bielinski et al., 2017; Gibson et al., 2017; Liko et al., 2020). We encourage that all stakeholders, including third-party administrators, policy makers, clinicians and PGx service providers, to jointly and regularly review existing evidence, in order to continue to establish clearer testing guidelines and expand insurance coverage for the wider adoption of PGx testing (Haidar et al., 2022; Keeling et al., 2019).

There are several strengths and weaknesses in this study. This innovative research serves as an encouraging step toward the clinical implementation of pre-emptive PGx testing in the outpatient setting in Asia, by laying the foundation for future research that can encompass a wider healthcare setting, such as tertiary hospitals. Furthermore, this study involved both primary and secondary care settings, allowing for a more comprehensive and integrated view of implementing PGx testing. One notable limitation of this study is its focus on private healthcare settings, which may restrict the generalizability of the findings to larger and more diverse populations.

Conclusion

It is highly feasible to implement a preemptive PGx testing in outpatient clinics within private healthcare settings in Singapore. This study establishes a framework for the clinical integration of PGx testing, encompassing key aspects such as testing protocols, training, education, EMR integration, and the potential clinical impact. Our early findings emphasize the clinical benefits of PGx testing, identifying a high prevalence of clinically actionable variants in almost all patients who were being genotyped with a clinically relevant multi-gene panel. Notably, differing perspectives between patients and clinicians on the cost as a barrier to PGx testing were observed, with cost being the least important factor for patients. We encourage that all stakeholders, including third-party administrators, policy makers, clinicians and PGx service providers, to jointly and regularly review existing evidence, and establish clearer testing guidelines and expand insurance coverage for the wider adoption of PGx testing.

Declarations

Supplementary Materials

Supplementary Material S1: Sample pharmacogenomics report

Supplementary Material S2: Survey Questions