Article
This article is part of a series of case studies on the use of NGS and digital tools to drive precision oncology in the Asia Pacific region. Scroll to the bottom of the article for the full list.
Cancer is a high-priority disease in Hong Kong, but although treatment success can be improved with next-generation sequencing (NGS), adoption of NGS has been limited. For most patients in the territory, it is only used in the second-line or after standard effective treatments for cancer have failed, even though it may be beneficial at earlier points in the diagnostic odyssey.
To give us an idea of the obstacles and opportunities for NGS in clinical practice, we spoke with Dr Lam Tai-Chung from the Department of Clinical Oncology in the Faculty of Medicine of The University of Hong Kong. His institution has had first-hand experience of landmark cases and outstanding treatment responses from using NGS in seemingly terminal patients. Alongside molecular tumour boards (MTBs), NGS-guided treatments have led to remarkable patient responses.
“NGS has proven its value in helping doctors to identify druggable targets and ensuring clinical benefits, which ultimately makes it cost-effective for both the patient and healthcare system,” says Dr Lam. “Physician behaviours have changed such that surgeons now believe in the impact of NGS enough to recommend it.”
Diagnostic challenges and realities
As a doctor who deals with brain cancers, Dr Lam finds few treatments are effective for difficult cancers like glioblastoma, making it critical to use NGS earlier, or in the first-line. In aggressive brain tumours, Dr Lam therefore performs NGS early in the clinical treatment pathway, before relapse.
This approach, however, imposes cost pressures on patients (in-house NGS panels are not available in Hong Kong and commercial panels are effective but sometimes expensive). Post-NGS treatment options may impose additional costs that become prohibitive in the absence of public or private reimbursement. For example, glioblastoma patients with crizotinib-treatable Met amplifications may need to purchase a 6-month course of crizotinib, yet just one month of the drug costs 20% more than the diagnostic test.
Moreover, not all patients will benefit from NGS. Those with difficult or rare brain tumours, such as glioblastomas, face uncertain outcomes after getting an NGS test result. Even if a druggable target is found, that drug may not be available in Hong Kong or responses to it may be unpredictable.
Some neuro-oncologists may be interested in using blood plasma circulating tumour DNA (ctDNA), which can be sampled from cerebrospinal fluid, for NGS testing in smaller gene panels (50 – 100 genes versus 300 – 400 genes). Dr Lam cautions, however, that blood plasma NGS may not be able to reflect the true tumour situation in neuro-oncology as such cancers only affect the central nervous system, are not systemic and can lead to CNS failures.
Logistical and technical challenges aside, liquid biopsies do offer a higher predictive value and accuracy in solid cancers such as those affecting the breast and colon, and Dr Lam expects that as the gene panels evolve further, the need for tissue biopsies in these cancers will decline.
The value of AI and digital tools
Doctors may not be able to fully read the most comprehensive and informative NGS reports, even though they impact whether patients are referred for germline mutation screening or genetic consultations. Such lapses have medical and legal consequences if doctors miss cancers in a patient’s offspring.
Digital analysis tools can minimise these risks and also help screen patients in a timely manner. Dr Lam hopes that these tools and platforms will soon include revised drug treatment recommendations, clinical trial updates or recent research on treatment options, even if off-label. Clinical practice would also benefit from the ability to use these tools to predict outcomes according to a patient’s indications or by ranking treatments according to projected response rates, costs or side effects.
Dr Lam emphasised that doctors need a protocol and artificial intelligence (AI)-driven pathway to provide clinical decision support on a patient’s initial diagnosis and throughout profiling. Doctors value having dynamic algorithms that can compile and analyse data to offer standardised next-steps, information on the different stages of disease and treatments, and education on implementing therapies.
As physicians start acquiring direct experience of NGS’ successes, it is hoped that these tools will quickly become a routine part of precision medicine in the clinic.
To learn more about the evolving role of next-generation sequencing, molecular tumour boards and clinical decision support tools in cancer care, check out some of these other case studies:
