Real-World Diagnostic Yield and Longitudinal Dynamics of Fluorescence In Situ Hybridization in Hematologic Malignancies: A Five-Year Retrospective Analysis of 1357 Patient Samples.
Cytogenetic patterns detected by FISH in contemporary hematology practice are strongly associated with clinical context, disease stage, and treatment history, with follow-up specimens showing significantly lower odds of abnormal findings and relapse samples demonstrating higher odds compared with diagnostic samples.
Key Findings
Results
Overall, 29.2% of informative FISH assays revealed cytogenetic abnormalities across the full cohort of hematologic malignancy patients.
The analysis included 1357 consecutive FISH assays performed between 2019 and 2023.
The assays were performed in 1092 patients with hematologic malignancies.
The cohort had a median age of 68 years.
The cohort was predominantly composed of patients with myelodysplastic/myeloproliferative neoplasms (27.2%), multiple myeloma (26.9%), and chronic lymphocytic leukemia (21.6%).
Results
Clinical indication was the strongest predictor of cytogenetic aberration detection in multivariable logistic regression.
Multivariable logistic regression was used to identify predictors of aberration detection.
Follow-up specimens showed significantly lower odds of abnormal findings compared with diagnostic samples (adjusted OR 0.48, 95% CI 0.36–0.64; p < 0.001).
Relapse samples demonstrated higher odds of abnormal findings compared with diagnostic samples (adjusted OR 1.82, 95% CI 1.28–2.59; p = 0.001).
These findings indicate that the clinical context (diagnostic, follow-up, or relapse) substantially shapes the likelihood of detecting a cytogenetic abnormality.
Results
Increasing age was independently associated with abnormal FISH results.
The adjusted odds ratio per decade of age was 1.24 (95% CI 1.12–1.38; p < 0.001).
This association was identified through multivariable logistic regression controlling for other predictors.
Age was an independent predictor separate from clinical indication and disease type.
Results
Several canonical cytogenetic abnormalities occurred less frequently in the real-world cohort than expected from historical diagnostic benchmarks.
BCR::ABL1 within the chronic myeloid leukemia testing pathway was observed less frequently than historical diagnostic series.
Del(13q14) occurred less frequently than expected in both chronic lymphocytic leukemia and multiple myeloma compared with published diagnostic cohorts.
The authors attribute this divergence to the fact that the real-world cohort included a mix of diagnostic, follow-up, and relapse samples rather than only treatment-naïve diagnostic patients.
The divergence highlights the need for context-specific interpretation of cytogenetic findings.
Results
Del(17p), involving the TP53 locus, remained consistently represented across disease categories and was enriched in chronic lymphocytic leukemia relative to published diagnostic cohorts.
Del(17p)/TP53 loss was observed across multiple hematologic malignancy disease categories.
The abnormality was enriched in chronic lymphocytic leukemia relative to published diagnostic cohorts.
The persistence of TP53 loss across disease boundaries was highlighted as underscoring the dynamic nature of clonal evolution.
This finding suggests TP53 loss may be selected for during treatment or disease progression.
Results
Longitudinal assessment revealed increasing cytogenetic heterogeneity over time in serially monitored patients.
Common cytogenetic lesions progressively declined over time in serially monitored patients.
Rare or complex abnormalities expanded from 16.3% to 36.7% of the observed cytogenetic landscape over the study period.
This progressive diversification of aberrations was interpreted as consistent with clonal evolution in hematologic malignancies.
Longitudinal analyses explored temporal changes in cytogenetic profiles among serially monitored patients specifically.
Discussion
The study's retrospective and cross-sectional design limits causal inference regarding the influence of treatment and disease stage on cytogenetic findings.
The authors note that the cross-sectional and retrospective design precludes inference of causal influence.
Most prior evidence supporting FISH interpretation originates from treatment-naïve diagnostic cohorts, providing limited insight into longitudinal dynamics.
The five-year retrospective analysis spanned 2019 to 2023 at a single institution.
The authors identified the need for context-specific interpretation of cytogenetic findings as a key practical implication.
What This Means
This research suggests that the results of a specialized genetic test called FISH (fluorescence in situ hybridization), used to detect chromosomal abnormalities in blood cancers, vary substantially depending on why and when the test is performed. Analyzing over 1,300 tests from more than 1,000 patients over five years, the researchers found that tests done during routine follow-up were about half as likely to detect abnormalities compared to tests done at initial diagnosis, while tests done when a patient's disease came back (relapsed) were nearly twice as likely to find abnormalities. Older patients were also more likely to have abnormal results, with the odds increasing about 24% for every decade of age.
The study also found that patterns of genetic abnormalities changed as patients were monitored over time. Common abnormalities became less frequent, while rare or complex ones grew from about 16% to nearly 37% of all detected abnormalities — a pattern consistent with cancers evolving and adapting, possibly in response to treatment. One particular abnormality, loss of a gene called TP53 on chromosome 17, was persistently detected across different types of blood cancers and appeared even more commonly in certain cancers (chronic lymphocytic leukemia) than historical studies would suggest, hinting that it may be selected for during disease progression or treatment.
This research suggests that published reference rates for how often FISH detects abnormalities — which are largely based on newly diagnosed, untreated patients — may not be directly applicable to real-world clinical practice, where patients are tested at many different points in their disease course. The findings highlight the importance of considering a patient's disease stage and treatment history when interpreting genetic test results in blood cancers, and suggest that tracking cytogenetic changes over time may offer insight into how these cancers evolve.
Fumo R, Scala K, Gaeta S, D'Ardia A, Infante T, Ciancia G, et al.. (2026). Real-World Diagnostic Yield and Longitudinal Dynamics of Fluorescence In Situ Hybridization in Hematologic Malignancies: A Five-Year Retrospective Analysis of 1357 Patient Samples.. Genes. https://doi.org/10.3390/genes17080955