Sequencing, Stratifying, and Standardizing

How did you come to focus on chronic myeloid leukemia?

When I first joined SA Pathology in 1997, I was tasked with developing quantitative molecular methods to assess residual disease in cancer. My first method measured the levels of BCR-ABL1 transcripts in patients with chronic myeloid leukemia (CML). The somatic BCR-ABL1 fusion is the primary genetic lesion that characterizes CML, and sensitive detection following stem cell transplant was a prognostic indicator. But to distinguish early relapse from residual disease, we needed a way to tell whether BCR-ABL1 levels were rising or falling. We used real-time quantitative PCR with hydrolysis probes – new technology at the time – and joined the first international clinical trial of a tyrosine kinase inhibitor, imatinib, in CML. With over 1,000 patients across three laboratories, the study soon demonstrated that imatinib therapy led to very rapid clearance of leukemia, and that monitoring at the molecular level provided prognostic information. That study kick-started my interest in the connections between treatment response kinetics, drug resistance, and patient outcomes.

How have patient outcomes changed since you entered the field?

They’ve changed dramatically! The disease is invariably fatal without therapy. When I entered the field, less than 40 percent of recently diagnosed patients qualified for transplant – a risky procedure that carried the danger of long-term morbidity or early mortality. The alternative, chemotherapy, was not great; interferon α could extend survival for the one-fifth of patients who responded well – but many couldn’t tolerate the side effects. It’s the introduction of imatinib and other potent tyrosine kinase inhibitors in recent years that has changed this once-fatal disease to one most patients survive long-term.

But there are still many challenges ahead. We don’t understand why some patients fail to respond to therapy or develop drug resistance; we can’t reliably identify patients who might benefit from more potent kinase inhibition despite the increased risk of cardiovascular events; we don’t know which patients might remain disease-free after drug cessation and which might relapse. Biomarkers measured at diagnosis may be the key to answering some of these questions – if we can determine which are most reliable, they may help us improve risk stratification, guide therapy choices, and provide patients and families with better peace of mind.

What’s your next research target?

We have used whole exome sequencing, whole transcriptome sequencing, and copy number variation to identify genomic variants in CML patients at diagnosis. This integrative genomics has revealed that patients with poorer outcomes exhibit a higher frequency of (potentially) clinically relevant variants – and those who progressed to acute leukemia also acquired similar variants. There is a substantial overlap between the variants we have detected in CML and those recently discovered in acute myeloid and lymphoid leukemias through next-generation sequencing, but there are also novel recurrent somatic variants that may be drivers of CML progression.

We aim to define the genomic mechanisms beyond BCR-ABL1 that are associated with treatment response, drug resistance, and early disease progression. The goal is to translate this new knowledge to the clinic by introducing a comprehensive biomarker testing panel at diagnosis that will reliably predict treatment response and guide decision-making. Integrative genomics may reveal the interplay between the different layers of genomic processes that regulate the kinetics of response and identify the most pathologically relevant genetic events.

You also lead international efforts for molecular method standardization…

Without standardization, it’s not possible to compare results across different labs. International recommendations now incorporate molecular response levels achieved at milestone time points over the first year of therapy to establish response. Patients who don’t reach these response levels are considered to have failed treatment and a change of therapy is mandated – but it’s hard to determine which patients are responding appropriately if we can’t compare one lab’s results with another’s.

Thanks to an extensive global effort, standardization has been largely successful – but the process has taken substantially longer than we initially anticipated. Molecular techniques are complex and many variables can impact the quality and consistency of final results. Standardized testing kits and newer, simpler technologies have aided the adoption of molecular monitoring for CML and benefited patients around the world. Our major challenge now is to introduce a reliable proficiency-testing program and – at the individual lab level – the routine use of affordable, appropriate quality control material.

Any advice for people starting out in molecular pathology?

Molecular pathology and genetic testing are moving rapidly forward. Because the volumes of data being generated are constantly increasing, I’d recommend that anyone seeking to enter the field prepare themselves with a thorough understanding of genetics and bioinformatics. I was lucky enough to hit the jackpot combination of fantastic guidance and mentorship, great opportunities, fascinating findings, a thirst for knowledge, supportive employers and colleagues, hard work, and serendipity. Some of those, of course, are luck of the draw – but others are readily available to those who look!