Nuts and Bolts of Consensus Recommendations of Biomarkers and Genetic Testing to Differentiate Behavioral Variant Frontotemporal Dementia From Primary Psychiatric Disorders

Sir,

Differentiating behavioral variant frontotemporal dementia (bv-FTD) from primary psychiatric disorders (PPD) continues to be a significant diagnostic challenge. A vital study on this aspect was published by Ducharme et al. ¹ The authors reviewed the existing literature in this article and made consensus recommendations for differentiating bv-FTD from PPD. Though the authors excellently brought out the diagnostic dilemmas, use of biomarkers, and genetic testing, a few points merit discussion.

The shared features between bv-FTD and PPD at clinical and genetic levels make it a challenging task to differentiate both accurately. Nearly 50% of bv-FTD receive a psychiatric diagnosis with a diagnostic delay of five to six years. ² Though this needs to be improved, factors such as the presence of frontal lobe features in PPD, FTD mimics, lack of reliable informants, poor cooperation to undergo structured assessments, co-existence of psychiatric symptoms in bv-FTD, past or family history of PPD and lack of adequate training during residency in dementia could have contributed for high diagnostic errors. Recently published diagnostic algorithms ³ and checklist tools ⁴ to clinically differentiate bv-FTD from PPD will help, to a certain extent, overcome the factors mentioned above.

Ducharme et al. ¹ recommended using biomarkers such as FDG-PET scan, SPECT scan, CSF tau, and CSF neurofilaments for early diagnosis of FTD. As the authors in the above article touched upon, most of the robust biomarkers are developed for Alzheimer’s dementia (AD). These biomarkers help in differentiating AD from FTD. Krudop WA et al., ⁵ showed that the specificity of FDG-PET to differentiate bv-FTD from PPD was shown to be poor in the follow-up study on the Late-onset frontal lobe cohort. ⁵ In this scenario, there is a concern about diagnosing bv-FTD in patients with biomarkers but meeting only subthreshold criteria (<3 out of six symptoms as per DSM-5 criteria) and minimal functional decline. In the context of AD, the new NIA-AA research criteria allow clinicians to diagnose patients presenting with biomarkers positive as Alzheimer’s disease and only be revised to AD if there are clinical features and functional decline. ⁶ Having similar terminology for FTD will also help identify the vulnerable or at-risk population without symptoms and monitor them over longitudinal follow-ups.

Another aspect is related to the use of genetic testing. The current recommendation is to use screening for three common congenital abnormalities (C9orf72, MAPT, GRN) if at least one of the first-degree relatives is affected with FTD. Ducharme et al. ¹ recommend genetic screening for late-onset PPD with a family history of FTD or ALS and in suspected bv-FTD with prominent psychiatric features or a family history of late-onset PPD. This recommendation is based on the finding of bv-FTD patients with C9orf72 repeat expansion to present with a prodrome of psychiatric diagnosis, GRN mutations with prominent visual hallucinations and delusions, and paranoid delusions in a few cases of MAPT mutations. The subsequent discussion predominantly focuses on C9orf72 to address some issues concerning PPD, clinical relevance, and prevalence of C9orf72 in FTD samples from India. In FTD samples with preceding psychotic symptoms, C9orf72 repeat expansions were reported in 21%–56%.^{7, 8} However, the evidence for a higher prevalence of C9orf72 repeat expansions in anxiety disorders, bipolar, or major depression is limited to a few reported case series. ⁹ Recommendation of C9orf72 mutation screening for late-onset psychosis with a family history of FTD/ALS or late-onset psychosis is justifiable, but currently not for other late-life PPD, which needs further investigation. Thus, genetic screening is also more relevant for late-onset psychosis with suspected FTD, as clinicians can exercise caution on the dose and duration of antipsychotics. In addition, variable interval in disease presentation of FTD, existence of minimally progressive FTD phenocopy, incomplete penetrance, and cut–off for repeat expansion that is considered pathological, which is not as robustly established as in Huntington’s disease are some of the concerns with C9orf72 screening. ⁹ There is a need for caution in this aspect as there is a risk of equating PPD or patients with non-specific frontal lobe features who screened positive for C9orf72 as bv-FTD. In addition, C9orf72 hexa-nucleotide repeat expansion has been reported to be rare in Indian FTD patients (7%). In lieu of earlier reports of the rarity of pathogenic MAPT and GRN mutations, it makes a strong case for the need to explore new genetic mutations in the Indian population. ¹⁰ In developing countries with resource constraints, C9ORF72 genetic screening (as well as for other genes: MAPT, GRN, TDP-43) needs to be based on the clinician risk assessment rather than a universal recommendation for all late-life psychiatry illnesses.

To conclude, prospective follow-up is recommended in cases presenting as late-onset PPD and subthreshold FTD/FTD mimics. The systematic application of appropriate tools and the availability of a well-trained multidisciplinary group can significantly increase diagnostic accuracy, especially in Indian settings, as facilities for novel diagnostics and genetic testing are sparse. ¹¹ Recommendations to include 3D MRI and genetic screening along with the CSF/serum biomarkers must be covered by government subsidies or as part of extensive population data bank studies.