Familial recurrence of amyotrophic lateral sclerosis (ALS) or frontotemporal dementia (FTD) has long been used as the clinical basis for assuming Mendelian, monogenic inheritance. That assumption informs choices about genetic testing, counselling, and gene-discovery strategies. However, both polygenic inheritance and rare pathogenic variants with reduced penetrance can produce familial clustering or apparently sporadic cases in small modern pedigrees, challenging the diagnostic value of a reported family history.
This study used simulation to quantify how well different family-history–based definitions of familial ALS predict the underlying genetic architecture, and to estimate rates of phenocopy and distant relatedness that might affect interpretation.
The authors developed APECS, a large-scale pedigree simulator that combines historical demographic parameters with unified frameworks for monogenic and polygenic disease models. APECS models three-generation pedigrees, ALS, FTD, and other dementias under alternate inheritance assumptions, enabling direct comparison of how family-history criteria map to underlying genetic architectures.
Key elements described in the source: APECS integrates demographic data and genetic models to simulate realistic family structures and disease occurrence. The simulator and analytic code are publicly available; the authors provided a GitHub repository and a web-based monogenic-probability calculator derived from their simulations.
The authors simulated 1,000,000 three-generation pedigrees, which produced 55.3 million synthetic individuals in total. From these simulated pedigrees they identified 5,431 ALS-affected index patients. Within this simulated index-patient group, 16.4% were modelled as having a monogenic basis for their disease.
Simulation results were benchmarked against real-world population estimates and were examined across a broad range of demographic and genetic parameters to test robustness of the findings.
The study evaluated common criteria for defining familial ALS (fALS) to determine their sensitivity and propensity for false classification when inferring monogenic disease.
Using the proposed criterion for 'definite fALS'—defined as two or more relatives with ALS among relatives within two degrees—the simulator found a sensitivity of 31.1% for detecting monogenic index patients. The same criterion also falsely classified 10.0% of cases labeled as 'fALS' as monogenic under the simulation parameters reported in the source.
These results indicate that standard, pedigree-based definitions of fALS miss a substantial fraction of monogenic cases (low sensitivity) while also producing false positives under some circumstances.
When relatives with definite FTD were also counted toward the 'definite fALS' definition, sensitivity rose to 43.3%, improving detection of monogenic index patients. However, when family-history criteria were broadened to include any type of dementia (unspecified dementias beyond definite FTD), the false positive rate increased markedly to 35.9%.
Thus, adding definite FTD to family-history criteria improved the ability to detect monogenic disease, but including unspecified dementias reduced the accuracy and increased the risk of misclassification.
The simulations also estimated a phenocopy rate among monogenic index patients. A phenocopy was defined as relatives who had ALS due to polygenic causes or who carried a pathogenic allele from a different ancestor. The reported phenocopy rate for monogenic index patients was 5.3% for relatives up to the third degree.
This non-negligible phenocopy rate implies that some affected relatives in a pedigree may not share the same causal monogenic variant as the index case, further complicating family-history interpretations and variant segregation assumptions.
Among monogenic index patients who appeared sporadic (no clear close relatives with ALS/FTD), the simulations found that 83.1% nevertheless had distant relatives with ALS or FTD up to the fifth degree. This suggests that searching for more distant relatedness can reveal familial aggregation that is not apparent on routine three-generation pedigree ascertainment.
Systematic methods to detect distant relatedness, in addition to broad genetic testing, may therefore help to identify monogenic cases that would otherwise be missed by conventional family-history criteria.
The main conclusion from the simulations is that a reported family history provides limited accuracy for inferring whether ALS in an index patient is monogenic versus polygenic. Relying solely on traditional fALS criteria will both miss monogenic cases that might benefit from genetic testing and targeted therapies and falsely alarm families when clustering is due to polygenic risk or phenocopies.
The authors recommend routine genetic testing irrespective of family history and advocate systematic approaches to detect distant relatedness as strategies to improve identification of known and novel genetic causes of ALS. Such approaches aim to reduce missed diagnoses and misclassification attributable to the limits of pedigree-based inference.
The APECS simulator code and data used in this study are available on GitHub, and the authors provided an online monogenic-probability calculator derived from their simulations.
Competing interests disclosed: one author reports sponsored research agreements with Biogen, Eli Lilly, Trace and AstraZeneca. Funding sources declared include the ALS Foundation Netherlands and the European Research Council (Grant agreement n° 772376 – EScORIAL).
Note: This work is a preprint posted to medRxiv and has not yet been peer reviewed; the authors and medRxiv state that the findings should not currently be used to guide clinical practice. The rewritten content above reports only findings and details provided in the source document.