Quality Improvement Project Enhances AML Genomic Testing Efficiency

Implementing a flow cytometry–triggered genomic testing algorithm and automating laboratory workflows can significantly reduce turnaround times for key molecular diagnostics in patients with newly diagnosed acute myeloid leukemia (AML), according to a quality improvement project. The study, published in JCO Oncology Practice, found the enhancements facilitated faster, more informed treatment decisions and supported personalized therapy and improved patient care.

The algorithm and automated workflows significantly cut turnaround times for genomic testing for newly-diagnosed acute myeloid leukemia (AML), enabling faster, more personalized treatment decisions.

Image Credit: RFBSIP - stock.adobe.com

As AML classification has shifted toward identifying genetically distinct subtypes, targeted therapies have been incorporated during induction treatment, which is ideally started within 5 days of diagnosis. For instance, the antibody-drug conjugate gemtuzumab ozogamicin (Mylotarg) may be added to induction chemotherapy, based on data from randomized controlled trials that showed a 5-year overall survival of 76.3% for patients who received gemtuzumab ozogamicin vs 55.2% for patients who did not.²

Early initiation of induction treatment was possible when genomics did not influence treatment decisions. “Since contemporary induction selection is guided by genomics, treatment initiation can be delayed while awaiting results,” the authors noted.

This quality improvement initiative evaluated if a flow cytometry–triggered diagnostic testing algorithm for newly diagnosed AML could enhance genomic test ordering and reduce turnaround times. The study took place at the London Health Science Centre in Ontario, Canada. The center treats and tests approximately 100 patients with newly diagnosed AML annually.

The researchers generated 3 ideas to improve test ordering and reduce turnaround times that included educating clinicians and laboratory personnel on contemporary guidelines and management of newly diagnosed AML, establishing a flow cytometry–triggered genomic diagnostic testing algorithm, and automating the laboratory workflow for next-generation sequencing.

They created a multidisciplinary core working group of 2 hematologists, 3 laboratory scientists, laboratory technologists, and 1 hematopathologist. After reviewing the clinical guidelines and AML therapies in Canada and identifying the target turnaround times for each diagnostic test, they developed a genomic testing algorithm. Subsequently, an educational intervention was organized in which a large multidisciplinary group of hematopathologists, molecular geneticists, and medical laboratory technologists was brought together to review guidelines and management of newly diagnosed AML, as well as the algorithm. Finally, the algorithm was implemented on May 18, 2023, approximately 5 months after the educational intervention.

The algorithm was activated in new patients with AML younger than 75 years of age, as these are the patients most likely to be eligible for intensive chemotherapy. To successfully activate the algorithm, marrow samples are collected, and tests such as morphology, flow cytometry, next-generation sequencing, and karyotyping are ordered.

In the final cycle, methodological changes were implemented that allowed the next-generation sequencing workflow to be automated.

At baseline, the frequency of test ordering and mean turnaround times were:

• 85.9% and 20.1 days for karyotyping
• 76.1% and 6.1 days for FLT3-ITD/TKD
• 78.9% and 31.1 days for next-generation sequencing
• 33.8% and 18.8 days for NPM1
• 11.2% and 12.0 days for RUNX1::RUNX1T1
• 7.0% and 7.6 days for CBFB::MYH11

After the algorithm was implemented, ordering improved to more than 90% compliance with guidelines for all the diagnostic testing. Only NPM1 showed a significant reduction in mean turnaround time (18.8-9.0 days; P = .0198) after just the implementation of the educational. After the algorithm was fully implemented, there were also significant reductions in turnaround times for next-generation sequencing (31.1-22.5 days; P < .0001) and karyotype (20.1-14.2 days; P = .082).

After the methodological changes were implemented in cycle 3, turnaround times were reduced another 37% for NPM1 to 5.3 days and by 50% for next-generation sequencing to 11.4 days. Next-generation sequencing also started detecting for RUNX1::RUNX1T1 and CBFB::MYH11 rearrangements, which reduced turnaround times from 12.0 to 6.1 days and from 7.7 days to 6.1 days, respectively.

“The use of guidelines and education alone is often insufficient in altering practice and our study highlighted the limitations of an education intervention,” the authors noted. “It was implementation of our flow cytometry–triggered genomic diagnostic testing algorithm for [newly diagnosed] AML that was most impactful, leading to more consistent ordering of relevant diagnostic tests.”

The researchers will continue to improve the algorithm as they uncover more gaps. In addition, they noted that use of this algorithm will likely be limited to larger academic centers due to the cost of implementation and maintenance.

References

1. Ho JM, Deotare U, Qureshi A, et al. Patient-centered genomic diagnostic testing for AML: a quality improvement project. JCO Oncol Pract. 2025:OP2400776. doi:10.1200/OP-24-00776

2. AML treatment has shifted toward more personalized approach. AJMC®. August 17, 2024. Accessed June 11, 2025. https://www.ajmc.com/view/aml-treatment-has-shifted-toward-more-personalized-approach