Rapid fluorescence in situ hybridisation optimises induction therapy for acute myeloid leukaemia.

Abstract

Recent approvals of multiple new agents for the treatment of acute myeloid leukaemia (AML) have highlighted the need for rapid and precise diagnostic laboratory studies.1, 2 Although the same cytarabine and anthracycline-based (typically daunorubicin) induction chemotherapy regimen (7 + 3) was the mainstay of therapy for fit patients for decades,3, 4 there is no longer a one-size-fits-all approach to induction chemotherapy. Instead, the optimal induction regimen is now selected based on a combination of clinical and disease-related factors.5 For example, the addition of gemtuzumab ozogamicin (GO) to standard induction chemotherapy improves outcomes in patients with core-binding factor (CBF) AML,6-9 while liposomal daunorubicin/cytarabine improves survival compared to standard 7 + 3 induction chemotherapy in older patients with defined subsets of AML with myelodysplasia-related changes (AML-MRC) and therapy-related AML.10 The defined subsets of AML-MRC that have been shown to have improved outcomes with liposomal daunorubicin/cytarabine are those with a known history of MDS or WHO-defined MDS-related cytogenetic changes. Our standard diagnostic pathway for patients with newly diagnosed AML includes morphology, immunophenotyping, metaphase chromosome analysis, rapid single-gene (FLT3 and IDH1/2) testing, reverse transcription polymerase chain reaction (RT-PCR) and/or fluorescence in situ hybridisation (FISH) for common leukaemia-associated chromosomal rearrangements, and a custom high-throughput sequencing panel which includes 68 genes recurrently mutated in haematologic malignancies. A modified pathway (with rapid PML-RARA FISH) is pursued in those with a high index of suspicion for acute promyelocytic leukaemia (APL). In order to ensure timely decision-making regarding the optimal induction therapy regimen for each patient, we have recently incorporated two rapid FISH panels into this diagnostic pathway (Fig 1). For all patients with a new diagnosis of AML other than APL, who are potential candidates for induction therapy, rapid CBF FISH for RUNX1/RUNX1T1 (associated with t(8;21)(q22;q22)) and CBFB translocation (associated with inv(16)(p13q22)/t(16;16)) using the RUNX1T1/RUNX1 dual colour–dual fusion (XL t(8;21) plus Translocation/Dual Fusion Probe, MetaSystems USA, Newton, MA, USA) and CBFB break apart (XL CBFB Break Apart Probe, MetaSystems USA) probes is performed to identify patients with CBF AML for whom the addition of GO to 7 + 3 induction therapy may be most beneficial. In addition, for patients with newly diagnosed AML between the ages of 60 and 75 we perform rapid FISH for the most common MDS-defining cytogenetic changes11 using probes for EGR1(5q31)/5p15.2 (XL Del(5)(q31) Deletion Probe, MetaSystems USA), D7S486(7q31)/7p11-q11 (CL 7q31 (D7S486) Deletion Probe, MetaSystems USA) and TP53(17p13.1)/NF1(17q11.2) (XL TP53/NF1 Deletion Probe, MetaSystems USA) (MDS FISH). The vast majority of AML samples that undergo MDS FISH testing also receive CBF FISH, given the less stringent age restrictions for the latter. Both rapid FISH studies are performed on peripheral blood or bone marrow specimens and utilise a short hybridisation time, in combination with optimised workflow in order to deliver results within 5·5 hours of specimen receipt in the CytoGenomics Laboratory, in comparison to the 5–7-day turnaround time for karyotype analysis at our institution. Samples first undergo direct harvest, with the cells pelleted and resuspended in hypotonic solution, followed by fixation and dropping onto slides using standard protocols. The appropriate FISH probes are then added to the slides, denatured and renatured over a total of 1·5 hours, and visualised. Also central to our rapid turnaround time is open and frequent communication between the laboratory, the treating physician and haematopathology to ensure that the laboratory is ready to receive samples and prioritise their evaluation. Here, we report our institutional experience with the use of our rapid FISH panels between 1 September 2017 and 31 July 2019. This analysis was approved by the University of Pennsylvania Institutional Review Board. A total of 116 patients with newly diagnosed non-APL AML had rapid CBF and/or MDS FISH performed during this time period. This included 50 patients with CBF FISH testing alone, 50 patients with both CBF and MDS FISH testing, and 16 patients with MDS FISH testing alone, with a total of 100 CBF and 66 MDS FISH tests thus being performed. The karyotype and FLT3 variant status are summarised in Table SI. During the 23-month period following the 2017 re-approval of GO, rapid CBF FISH was positive in 16 of 100 patients tested (16·0%) (Table I). The rapid CBF FISH panel identified all patients with inv(16) or t(8;21) by conventional karyotype, with no false-positive or false-negative results. All 14 patients who were determined to be eligible for induction therapy received GO in combination with 7 + 3. Over the same period, we performed the rapid MDS FISH panel on 66 patients, with positive results in 16 patients (24·2%) (Table I). Of these 16 patients, 14 had karyotypes that later confirmed the FISH findings. Samples for the remaining two patients failed to grow and karyotyping could not be performed. Of the 16 patients with AML-MRC as identified via our rapid MDS FISH panel, eight (50·0%) were treated with liposomal daunorubicin/cytarabine, six patients were treated with less intensive therapy due to either comorbidities or the preference of the patient and/or treating physician, and two were treated at other institutions. The panel had a sensitivity of 100% for del7/del(7q), del(5q) and TP53 deletion/monosomy 17. The panel of course failed to identify other MDS-associated cytogenetic abnormalities as defined by the WHO 2016 Classification in three patients, for an overall sensitivity of 84% and specificity of 100%. Of these three patients, one had a karyotype with the MDS-defining del(11q) and two had complex karyotypes (Table SII). An additional fourth patient with AML-MRC was diagnosed solely based on the presence of multilineage dysplasia with a normal karyotype. Of note, patients with this morphologically defined subset of AML-MRC are ineligible for liposomal daunorubicin/cytarabine. An alternative approach to rapid FISH could be to use direct culture methods in order to obtain full karyotype results within 18–24 hours of specimen receipt. However, with the personnel constraints and high volume of specimens in our laboratory we felt that it was prudent to dedicate resources to rapid FISH analysis, which has a substantially faster turnaround time of 5·5 hours. Furthermore, prior studies suggest that shorter culture time for karyotype analysis may result in a lower likelihood of detecting a chromosomal abnormality.12 There are several potential benefits of implementing rapid FISH panels to a diagnostic workflow. First, this allows timely decision-making regarding the optimal induction therapy regimen for each patient. For many patients this may result in a decreased length of initial hospitalisation. In our cohort, there was a median time from admission to beginning induction chemotherapy of 2·5 days for patients with positive CBF FISH (range 1–7 days), four days for patients with positive MDS FISH (range 2–7 days) and 3·5 days for the remaining patients (range 0–9 days). At our hospital the vast majority of patients who are hospitalised at initial diagnosis are not stable enough to be discharged prior to induction chemotherapy and a waiting period of 5–7 days for conventional cytogenetics would require either delay in therapy, lengthening of hospital stay, or initiation of therapy without full information and therefore choosing therapy that may not be optimal. While earlier initiation of induction chemotherapy has been reported to improve overall survival for some patients,13 other data suggest otherwise.14, 15 Additionally, knowing the results from the rapid FISH testing before other laboratory assays are completed can be valuable. FISH results often assist in elucidating complex conventional cytogenetics and can aid laboratorians in providing a more accurate description of the abnormal chromosomes. Similarly, detection of a patient who is positive for CBF AML rearrangement by rapid FISH, for example, may reduce the urgency for RT-PCR testing. Fusion breakpoint identification could be reported at a later time point for the CBF FISH-positive patients. At our institution, we incur a greater laboratory expense by performing the rapid FISH panels. Although the laboratory costs are increased, rapid identification of these patients and subsequent initiation of the appropriate induction therapies may reduce other health care costs through reduction of inpatient length of stay, although a formal cost–benefit analysis is beyond the scope of this paper. In summary, we report the implementation of two rapid FISH panels used to inform decisions on induction chemotherapy regimen for patients with newly diagnosed AML. The results from the MDS and CBF FISH panels are available within 5·5 hours of specimen receipt, and all panels use commercially available FISH probes. Our results demonstrate the utility of these rapid FISH panels in identifying patients with CBF AML (for whom GO should be considered) or patients with newly diagnosed AML who have MDS-associated cytogenetic abnormalities (for whom liposomal daunorubicin/cytarabine should be considered). This approach has shown us the value of using rapid FISH for the early diagnosis and management of newly diagnosed AML, and our institution plans to continue performing rapid CBF and MDS FISH as part of our standard AML diagnostic evaluation, with potential inclusion of additional probes as more genetically driven agents are approved for AML therapy. The authors thank Vania Aikawa M.D. and Honore Strauser for their assistance. NDN and CMM collected the data and wrote the manuscript. FEN and CF contributed to the acquisition of the data. JJR, AB, SML and JJDM contributed to acquisition of the data, FISH panel design and oversaw the project. AB has served as a paid consultant on an AML advisory board for Jazz Pharmaceuticals, producers of liposomal daunorubicin/cytarabine. The other authors have no conflict of interest to declare. Table SI. Clinical characteristics of the de novo AML patients included in this study. Table SII. Karyotypes of patients with first-time diagnosis of AML-MRC that were not identified by rapid MDS FISH. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.