Introduction
Before converging on neurofilaments and cryptic peptides as the foundation for our biosensor, we carried out an extensive exploration of different molecular candidates known to be altered in amyotrophic lateral sclerosis (ALS).
Our guiding principle was to prioritize biomarkers that could be detected at early stages of the disease, that held potential for translation into a clinical setting, preferably using blood samples.
While several options showed scientific promise in research studies, each carried limitations or advantages that made them more or less suitable for our final design. What follows is a summary of the main candidates we evaluated, the reasons why they initially seemed attractive, and the ones why they were ultimately set aside or selected.
Initial Candidates
1. Natural Killer Cell Subpopulations
One of the first avenues we examined involved immune cells, specifically two subpopulations of natural killer (NK) cells that have been reported to change in ALS patients. These subsets are easily measurable in peripheral blood and were correlated with known markers of neuronal injury, making them appealing for non-invasive detection. Biologically, the distinction between the two subsets lies in their level of maturation and function: one population is considered more cytotoxic, while the other is more regulatory.
Despite these interesting features, the evidence base remained thin. The available studies involved small patient cohorts and lacked comparisons across other neurodegenerative conditions. This meant that, although the signal was clear in ALS, it was impossible to know whether it was specific to this disease or shared with others. For these reasons, we decided not to pursue NK cells as part of our detection strategy for the moment.
2. Circulating microRNAs
We also investigated circulating microRNAs released from neuronal vesicles in the blood. These short non-coding RNAs regulate gene expression and have been proposed as highly sensitive indicators of neurodegeneration. In certain pilot studies, panels of microRNAs achieved near-perfect sensitivity and specificity in distinguishing ALS patients from controls. Their non-invasive nature, reliance on plasma samples and apparent diagnostic accuracy make them a highly promising diagnostic tool.
However, this promise came with important caveats. The findings were not yet validated in longitudinal cohorts, nor in patients at early stages of the disease, and reproducibility across different laboratories remained uncertain. From an entrepreneurial perspective, microRNA detection is a well-established, standardized process (typically qPCR), making it difficult to secure intellectual property rights and therefore harder to attract funding or develop a differentiated commercial product. Furthermore, microRNA expression patterns are dynamic and may overlap with other neurological or systemic conditions, undermining specificity. Given these challenges, we believe that further research is necessary before considering its potential use as a diagnostic biomarker.
3. Muscle-Related microRNA
We also looked into a microRNA closely linked to muscle regeneration and neuromuscular plasticity. Its expression levels were consistently higher in ALS patients, and it correlated with both neurofilament concentrations and clinical progression scores. Importantly, it had been suggested as a potential early marker, as its upregulation appears before advanced symptoms develop.
Nevertheless, the specificity issue emerged once again. This microRNA is not unique to ALS but is also altered in other muscle-related diseases such as dystrophies and inflammatory myopathies. From a technical standpoint, the most reliable detection required isolating exosomes of neuronal origin, adding a level of complexity incompatible with a simplified biosensor design. Although it is valuable for understanding disease mechanisms and studying progression, its limitations in precision and practicality led us to exclude it from consideration.
4. Neurofilaments
Neurofilaments constitute one of the principal protein components of the neuronal cytoskeleton, playing a critical role in preserving the structural integrity and functionality of neurons. They consist of 4 main proteins: Neurofilament light chain (NfL), neurofilament medium chain (NfM), neurofilament heavy chain (NfH) and α-internexin.
NfL is the most abundant neurofilament subunit and can be detected in both blood and cerebrospinal fluid (CSF). Its measurement is already being implemented in clinical practice as a biomarker of global neurodegeneration. Numerous studies have consistently shown that both NfL pNfH concentrations are significantly elevated in ALS patients compared with healthy controls and individuals with other neurological diseases. Quantification provides valuable information: higher baseline concentrations of NfL/pNfH consistently correlate with shorter survival and a more aggressive disease course.
These markers are valuable not only for early diagnosis, but also for prognosis and for monitoring disease progression. They are among the most established and validated biomarkers of neurodegeneration in ALS. Their detection is minimally invasive and considerably less expensive than previous approaches, such as magnetic resonance imaging, which has accelerated their integration into routine clinical practice.
However, neurofilaments are not specific to ALS; they also increase in other neurodegenerative diseases or acute neurological injury. Therefore, they are generally considered indicators of neuronal damage.
Considering that this biomarker is widely validated and already in clinical use, we have decided to include it in our ALS diagnostic panel. Our mid-term objective is to develop a more sensitive and efficient detection system than the current standard. However, because the existing detection method is already clinically effective, we have chosen to prioritize experimental work on other biomarkers that may provide higher impact and innovation opportunities in early diagnosis.
5. Cryptic peptides
TAR DNA-binding protein 43 (TDP-43) is a nuclear protein essential for maintaining RNA integrity by repressing the inclusion of non-functional "cryptic" exons. In the vast majority of ALS cases, as well as in approximately half of frontotemporal dementia (FTD) cases, together referred to as the TDP-43 proteinopathies, the nuclear loss and cytoplasmic aggregation of TDP-43 disrupt this regulatory function. As a result, exons are erroneously incorporated into mature mRNA – so-called "cryptic exons" – resulting in aberrant protein isoforms absent in healthy tissue. It is considered a pathological hallmark in ~97% of ALS cases.
These cryptic isoforms are highly specific to ALS and FTD, consistently correlate with ALS pathology and also can be detected prior to clinical neurodegeneration. Although they are shared by both diseases, the symptoms of ALS or FTD clearly differ already in its early stages. While ALS affects mainly the movement of the body, FTD affects more the cognitive function. The cryptic peptides can be quantified from blood or CSF, making them promising biomarkers for early ALS diagnosis.
Their direct association with neuronal dysfunction, emergence as a result of well-characterized molecular mechanisms and their presence correlates consistently with ALS pathology. Importantly, their biochemical properties make them better suitable for synthetic biology approaches, where engineered recognition elements can be designed to detect them with high specificity.
In view of the lack of available standard methods for its detection in clinical routine and all the positive technical aspects, we decided to support our diagnostic system primarily on them.
6. Retroviral proteins
Human endogenous retrovirus K (HERV-K) is a group of retroviral genes integrated in the human genome. Some of its proteins are aberrantly expressed in cancers, autoimmune diseases and neurodegenerative diseases, although its exact role remains unclear. It has been shown that HERV-K is activated in a subpopulation of patients with sporadic ALS and that specific proteins may contribute to neuroinflammation and neurodegeneration.
From this observation, clinical trials started with the aim of using antiretrovirals for the treatment of ALS. Nevertheless, no efficacy was observed. It remains unclear whether it was due to the intrinsic inefficacy of the treatment or due to a lack of patient stratification. One plausible hypothesis is that antiretrovirals can only be effective in patients that express HERV-K.
Considering such a need for the detection of HERV-K proteins, their detectability in blood, and the lack of standard methods of detection, we evaluated the potential of developing a diagnostic kit for HERV-K proteins. Although we haven't discarded to do it in future stages of the project, the limited validation of HERV-K presence in early disease stages and the lack of consistent findings across patient groups led us to prioritize other biomarkers for the current phase of our work.
While viral envelope peptides were an interesting conceptual candidate, we ultimately prioritized cryptic peptides as the more reliable and feasible choice for our project.
Selected Biomarkers: Neurofilaments and Cryptic Peptides
After carefully weighing the reported options, we decided to focus on cryptic peptides and to complete the diagnostic panel with neurofilaments. Together, they meet key requirements: appearing at different stages of progression, being linked to neuronal pathology and being measurable in blood.
The added value lies in their complementarity. Neurofilaments are a structural marker, widely validated for reporting overall neuronal damage, while cryptic peptides are a functional marker, capturing ongoing molecular processes for TDP-43 pathies. By combining them, we create a multimodal panel that increases both sensitivity and specificity.
Cryptic peptides have a distinct temporal profile compared with traditional neurofilaments. Evidence from comparative studies shows that cryptic peptide levels increase in early stages (sometimes even before symptom onset) and subsequently decline in advanced stages, likely reflecting the loss of viable neurons with functional TDP-43. Neurofilament markers remain relatively low during the presymptomatic phase and rise later, once neurodegeneration becomes clinically apparent.
This complementary pattern suggests a dual diagnostic role: while neurofilaments primarily reflect established neuronal damage, cryptic peptides capture the earliest phases of pathology, preceding widespread degeneration. For this reason, TDP-43 cryptic peptides have been proposed as early diagnostic biomarkers of ALS, capable of signaling the core molecular defect before clinical presentation.
Combining cryptic peptides with neurofilament measurements may enhance prognostic accuracy. Indeed, one study indicates that assessing both cryptic peptides and NfL could refine patient stratification by disease stage and aggressiveness. This approach is especially relevant in ALS, a condition known for its clinical and biological heterogeneity. With our panel, we can achieve earlier diagnosis, even before motor symptoms appear, monitor progression over time, providing information for personalized management.
Finally, our biosensor is not meant to replace neurological diagnosis but to complement clinical criteria. It provides a molecular tool to support decision-making, enable earlier interventions, improve patient stratification, and contribute to a better understanding of ALS trajectories.
Final Conclusions - Summary
Considering the biomarkers above, we got to the following conclusions:
- NK subpopulation detection may be promising, but lacks evidence of its effectiveness for an early ALS diagnostic.
- Using a set of microRNAs may be a good way to diagnose ALS. It lacks studies in early stages of the disease. Person-to-person variability risk. The detection method would be very standard methods, therefore difficult to patent.
- The current system for the detection of neurofilaments is implemented, but could be improved. It is not an urgent medical need.
- The detection of the retrovirus in blood may be critical to validate the efficacy of antiretrovirals for the treatment of neurodegenerative diseases. Remains a good candidate, but the lack of solid studies of its early presence in blood favored the choosing of cryptic peptides as our candidates for the diagnosis.
- The protein isoforms derived from the expression of cryptic exons due to the inactivation of TDP-43 (above referred as cryptic peptides) have been clinically observed in early stages of ALS and FTD. The initial symptoms of these two diseases are clearly differentiable. Therefore, a comprehensive observation of clinical traits and the presence of cryptic peptides would be specific for ALS. There is no standard method to detect them, and the system used to detect its presence can be strongly patented.
References
- Álvarez-Sánchez, E., Carbayo, Á., Valle-Tamayo, N., Muñoz, L., Aumatell, J., Torres, S., Rubio-Guerra, S., García-Castro, J., Selma-González, J., Alcolea, D., Turon-Sans, J., Lleó, A., Illán-Gala, I., Fortea, J., Rojas-García, R., & Dols-Icardo, O. (2025). Single-cell RNA sequencing highlights the role of distinct natural killer subsets in sporadic amyotrophic lateral sclerosis. Journal of Neuroinflammation, 22(1), 15.
- Banack, S. A., Dunlop, R. A., Mehta, P., Mitsumoto, H., Wood, S. P., Han, M., & Cox, P. A. (2024). A microRNA diagnostic biomarker for amyotrophic lateral sclerosis. Brain Communications, 6(5), fcae268.
- Dong, S., Liu, X., Zhou, Y., Li, J., Qi, Z., Wang, Z., Yang, W., & Chen, X. (2025). Prognostic Value of Cerebrospinal Fluid and Serum Neurofilament Light Chain in Amyotrophic Lateral Sclerosis: A Correlation Study. Brain and Behavior, 15(1), e70256.
- Feneberg, E., Gray, E., Ansorge, O., Talbot, K., & Turner, M. R. (2018). Towards a TDP-43-Based Biomarker for ALS and FTLD. Molecular Neurobiology, 55(10), 7789–7801.
- Irwin, K. E., Jasin, P., Braunstein, K. E., Sinha, I. R., Garret, M. A., Bowden, K. D., Chang, K., Troncoso, J. C., Moghekar, A., Oh, E. S., Raitcheva, D., Bartlett, D., Miller, T., Berry, J. D., Traynor, B. J., Ling, J. P., & Wong, P. C. (2024). A fluid biomarker reveals loss of TDP-43 splicing repression in presymptomatic ALS-FTD. Nature Medicine, 30(2), 382–393.
- Seddighi, S., Qi, Y. A., Brown, A.-L., Wilkins, O. G., Bereda, C., Belair, C., Zhang, Y.-J., Prudencio, M., Keuss, M. J., Khandeshi, A., Pickles, S., Kargbo-Hill, S. E., Hawrot, J., Ramos, D. M., Yuan, H., Roberts, J., Sacramento, E. K., Shah, S. I., Nalls, M. A., … Ward, M. E. (2024). Mis-spliced transcripts generate de novo proteins in TDP-43-related ALS/FTD. Science Translational Medicine, 16(734), eadg7162.
- Verde, F. (2025). Neurochemical biomarkers of amyotrophic lateral sclerosis: recent developments. Current Opinion in Neurology, 38(5), 614–619.
- Zheng, Q., Wang, D., Lin, R., Chen, Y., Huang, H., Xu, Z., Zheng, C., & Xu, W. (2023). Mendelian randomization analysis suggests no associations of human herpes viruses with amyotrophic lateral sclerosis. Frontiers in Neuroscience, 17, 1299122.