Reducing Cytogenetic Testing in the Era of Next Generation Sequencing (NGS); Are We Choosing Wisely?
Bibliographic record
Abstract
Background: The combination of automation and expanding panel of target genes has improved utility and reduced costs of Next Generation Sequencing (NGS), leading to its widespread adoption in the clinical laboratory. In most laboratories, NGS has been added without consideration for redundancy or relative value compared to traditional genomic assays such as G-band karyotyping and FISH. At our centre, most patients with suspected hematologic malignancies receive both conventional cytogenetics (CG) and NGS assessment in addition to bone marrow morphology and flow cytometry. Appropriate test utilization is a high priority highlighted by campaigns such as Choosing Wisely, which often disproportionally focus on appropriate utilization of "routine" high volume tests rather than new test modalities. We implemented NGS screening in patients with suspected hematologic malignancies (Levi et al. 2019), and demonstrated enhanced diagnostic and prognostic yield of NGS, supporting the efficacy and cost-effectiveness of an 'NGS first' approach with CG restricted to samples with morphologic abnormalities in MDS (Kawata et al. BJH 2020). In this follow up study, we further expanded our Morphologic Flow Triage/NGS first (MFT-NGS1) algorithm to investigate patients with suspected hematological diseases. The main objective of our study was to evaluate this rationalized molecular diagnostic testing "MFT-NGS1" algorithm for its feasibility, acceptability and cost impact. Methods: Using the results from morphologic interpretation of aspirate and flow cytometry, patient samples were triaged into 4 groups. Group 1: Patients with dysplastic features in the marrow or excess blasts were triaged to Bone Marrow Molecular Diagnostic 1 (BMD1) and had both NGS and G-band karyotyping. Group 2: Patients with no excess blasts or dysplasia (BMD2), had NGS only with CG sample held for 3 months in case testing was required in follow up. Group 3: Patients who had NGS and/or CG on a previous BM aspirate were triaged to (BMD3), with the comment that NGS and CG should not be repeated unless results will influence patient management. These samples were held for 4 weeks and testing was performed only if specifically requested. Group 4: Patients with suspected myeloma where the proportion of plasma cells was less than 5% were triaged to (BMD4), and FISH testing was cancelled. These samples were held for 3 months in case subsequent biopsy showed an increased proportion of plasma cells in keeping with myeloma. Results: Over a 9-month period between August 2019 to April 2020 a total of 599 adult BM samples were assessed; 549 (91.7%) were ordered by hematologists and 331 (60.1%) meeting study criteria for MFT-NGS1 algorithm. Of those, 115 (34.7%) samples showed morphologic abnormalities and triaged to BMD1; 61 (18.4%) samples showed no morphologic abnormalities and were triaged to BMD2; 116 (35%) had previous CG and/or NGS tested and triaged to BMD3; and 39 (11.8%) samples were from patients with suspected myeloma had less than 5% of plasma cells and were triaged into BMD4 group. (Figure 1) Overall the MFT-NGS1 algorithm decreases G band karyotyping or FISH analysis in 149/331 (45%) samples. Only 11 / 216 (5.5%) hematologist overruled triage comment and requested CG testing for a specific indication either suspected progression of known diseases or for a therapeutic decision (e.g. drug approval). CG and FISH were mistakenly performed without the necessary reconfirmation in 26/216 (12%). Conclusion: The proposed approach combining morphologic and flow triage and NGS as the primary genomic test (MFT-NGS) is both feasible and well accepted by clinical teams. We estimated that approximately 40% of all CG testing could be reduced primarily by reducing repeat testing which offsets a large proportion of the cost of implementation of NGS testing, while increasing both diagnostic and prognostic yield. Key factors for the success of this quality improvement project were involvement of clinical and genomic teams in developing the triage algorithm, rapid turnaround time (less than 24 to 48 hours) for aspirate interpretation and flow for triage and communication between laboratories within a fully integrated healthcare centre. Figure 1 Disclosures No relevant conflicts of interest to declare.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.020 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".