Veristat announced on November 12, 2026, that it has partnered with Clene Nanomedicine, a subsidiary of Clene Inc. (Nasdaq: CLNN), to accelerate biostatistical analyses supporting the New Drug Application for CNM-Au8, an investigational treatment for amyotrophic lateral sclerosis. Using its InStat automated analytics platform, Veristat reported reducing the time required to produce validated statistical tables, listings and figures from as much as six weeks to approximately one week, representing an estimated 80% improvement in the relevant workflow.
The collaboration addresses a time-sensitive operational requirement for Clene, which submitted its CNM-Au8 application to the United States Food and Drug Administration in October 2026 seeking accelerated approval. The submission incorporates neurofilament light chain biomarker analyses and clinical information from several studies, including the HEALEY ALS Platform Trial, RESCUE-ALS and associated longer-term follow-up programmes. However, the FDA has not yet confirmed acceptance of the application for substantive review in the company’s latest disclosed regulatory update.
The announcement is principally a clinical data processing milestone rather than a new efficacy result or regulatory decision. Its importance lies in demonstrating how automated statistical production may help smaller biotechnology companies prepare complex submission packages under tight development and financing constraints. Whether CNM-Au8 can ultimately secure approval remains dependent on the FDA’s assessment of the scientific evidence, particularly given the limitations of the original randomised trial results.
How did Veristat reduce Clene’s biostatistical analysis timeline?
Clinical trial regulatory submissions require large quantities of structured statistical information that allow regulators to examine study results, participant characteristics, safety findings and treatment outcomes. These materials commonly include tables, listings and figures generated according to statistical analysis plans and applicable regulatory data standards. Producing them can involve extensive programming, independent checking and reconciliation of study datasets.
Traditional workflows often rely on statistical programmers writing analysis code for individual outputs. Each table or figure must be checked against the relevant analysis specifications, while discrepancies may require additional programming and quality review. The process can become particularly demanding when sponsors need to incorporate late-stage analyses or modify the submission package.
Veristat’s InStat platform is designed to automate much of this production process. The system uses artificial intelligence-assisted tools to translate biostatisticians’ descriptions into structured analytical specifications, which are then executed through a library of validated statistical engines. Veristat describes the approach as avoiding reliance on large language models to generate the underlying statistical analysis code directly.
For Clene, the platform supported neurofilament light chain analyses and other regulatory deliverables associated with its CNM-Au8 filing. Veristat reported that the work involved twelve deliverables, each with its own statistical analysis plan. The system allowed the company to generate reviewable outputs while additional analyses were incorporated into the filing preparation process.
The reported reduction from as much as six weeks to one week represents a substantial operational improvement for the specific work described. However, this is a company-reported performance measure and should not be interpreted as an independently established industry-wide benchmark. The extent of time savings in another clinical programme would depend on dataset quality, analysis complexity, existing programming infrastructure and review requirements.
Veristat also stated that the resulting outputs remained traceable and subject to expert biostatistical review. These controls are essential because automation does not eliminate the need to verify that the correct populations, endpoints and statistical methods have been used. A faster calculation is only useful for regulatory purposes if the underlying analysis is scientifically appropriate and reproducible.
Why are statistical tables, listings and figures critical to an ALS New Drug Application?
A New Drug Application contains detailed evidence intended to support the safety and efficacy of a proposed medicine. The statistical analysis package enables regulators to evaluate the information underlying a sponsor’s conclusions rather than relying solely on summary statements. Tables, listings and figures provide structured presentations of outcomes, safety events, patient characteristics and related analyses.
In a conventional clinical development programme, these outputs can include analyses of prespecified primary and secondary endpoints, safety populations, treatment exposure and protocol deviations. They may also include sensitivity analyses intended to test whether conclusions remain consistent under different statistical assumptions. Each output must correspond to an appropriate dataset and clearly defined analytical method.
Clene’s application presents additional complexity because its proposed accelerated approval strategy relies heavily on neurofilament light chain, a biomarker associated with neuronal injury. The company is seeking to connect biomarker changes with clinical outcomes in amyotrophic lateral sclerosis, a progressive neurodegenerative disease affecting motor neurons.
The relationship between biomarker measurements and clinical outcomes is therefore central to regulatory interpretation. Researchers must establish whether observed changes are consistent across appropriate populations and how they relate to survival, respiratory function and other relevant outcomes. Statistical analyses can help characterise those relationships, but they cannot independently prove that the biomarker is a valid surrogate for treatment benefit.
Veristat’s automation may improve the efficiency with which these analyses are produced and reviewed. It does not change the underlying results or resolve limitations in the original clinical evidence. The FDA must still determine whether the submitted analyses support the proposed accelerated approval pathway.
What is the current FDA regulatory status of CNM-Au8?
Clene announced on October 19, 2026, that it had submitted a New Drug Application for CNM-Au8 seeking accelerated approval for amyotrophic lateral sclerosis. The application draws on data from the Phase 2 HEALEY ALS Platform Trial, the Phase 2 RESCUE-ALS study, open-label extensions and an expanded access programme supported by the United States National Institutes of Health. The filing represents a significant development milestone for a company whose lead candidate remains investigational.
Submission does not mean the FDA has accepted the application for substantive review. During the filing assessment, the agency determines whether the submitted package is sufficiently complete to permit a full review. Acceptance would initiate the formal review process but would not establish that the medicine is effective or that approval is likely.
Clene has indicated that it expects an FDA response concerning application acceptance before the end of 2026. Until that decision is disclosed, the regulatory pathway remains dependent on the agency’s assessment of the filing package. The existence of validated statistical outputs does not guarantee that the application will be accepted.
Accelerated approval also involves an important evidentiary distinction. The pathway can permit approval based on a surrogate endpoint reasonably likely to predict clinical benefit in appropriate circumstances, but the suitability of that endpoint and the overall evidence must be assessed for the specific medicine and indication. A biomarker with recognised prognostic value is not automatically an accepted surrogate for demonstrating treatment efficacy.
Clene believes its neurofilament light chain findings may support that pathway. However, the company must establish that treatment-related changes in the biomarker provide sufficient evidence for the proposed regulatory decision. The FDA’s review will consider the complete clinical dataset, the statistical analyses and the relationship between biomarker changes and patient outcomes.
What do the original HEALEY ALS and RESCUE-ALS trials establish about CNM-Au8?
CNM-Au8 is an investigational oral suspension containing catalytically active gold nanocrystals. Clene is developing it to influence cellular energy metabolism and neuronal function through mechanisms involving mitochondrial activity and nicotinamide adenine dinucleotide pathways. The therapeutic hypothesis is that supporting neuronal energy processes may help address aspects of neurodegeneration.
The biological rationale provides a basis for clinical investigation, but the original randomised studies did not establish confirmatory efficacy. The HEALEY ALS Platform Trial evaluated CNM-Au8 in a randomised, double-blind setting involving 161 participants assigned within the treatment regimen, with additional shared placebo data available for relevant analyses. The primary endpoint assessed disease progression over 24 weeks using a combined measure involving function and survival.
The peer-reviewed HEALEY results did not demonstrate a statistically significant improvement in the primary efficacy outcome. The reported disease rate ratio was 0.97, with a 95% credible interval of 0.783 to 1.175, and key secondary outcomes also failed to demonstrate a treatment benefit during the controlled period. These findings represent an important limitation when evaluating the current regulatory strategy.
Researchers subsequently examined exploratory survival outcomes and longer-term follow-up information. Some analyses suggested favourable survival patterns in selected treatment populations, particularly after adjustment for baseline differences. However, these findings were hypothesis-generating and did not overturn the negative primary trial result.
The RESCUE-ALS Phase 2 study also failed to meet its primary and key secondary endpoints. Exploratory analyses and longer-term observations identified potential signals of benefit, but these findings were subject to limitations involving statistical testing, sample size and the interpretation of open-label data. They cannot be treated as equivalent to a successful confirmatory clinical trial.
The distinction is central to the CNM-Au8 regulatory case. Clene’s filing relies on a broader evidence package that includes biomarker changes and clinical observations generated after the original randomised comparisons. Whether those analyses provide sufficient support for accelerated approval is a matter for FDA review rather than a conclusion established by the company’s submission.
Why is neurofilament light chain central to Clene’s accelerated approval strategy?
Neurofilament light chain is a structural protein associated with neurons that can be measured in biological fluids. Elevated levels may reflect neuronal injury and have been studied as indicators of disease activity or prognosis in several neurological conditions. In amyotrophic lateral sclerosis, neurofilament measurements are associated with disease severity and progression in relevant research populations.
Clene is seeking to use evidence involving neurofilament light chain to support the argument that CNM-Au8 may influence underlying neurodegenerative processes. The company has reported post hoc analyses in which participants showing reductions or stabilisation of neurofilament levels experienced more favourable clinical outcomes. These observations have contributed to its proposed accelerated approval strategy.
However, association between a biomarker and patient outcomes does not independently establish a treatment effect. Participants whose biomarker levels decline or remain stable may differ from others in baseline disease characteristics, progression rates or other factors. Analyses based on biomarker response after treatment can therefore be vulnerable to selection effects and confounding.
For regulatory purposes, the more demanding question is whether treatment causes a change in the biomarker that is reasonably likely to predict meaningful clinical benefit. This requires consideration of the biological mechanism, available clinical evidence and the consistency of findings across studies. A prognostic biomarker may provide useful information about disease progression without necessarily serving as a validated surrogate for drug efficacy.
Clene has stated that the FDA acknowledged the established prognostic value of neurofilament light chain during earlier regulatory discussions and recognised its potential relevance to an accelerated approval approach. That acknowledgement should not be interpreted as the agency having accepted CNM-Au8’s biomarker findings as sufficient for approval. The company must still satisfy the requirements applicable to its specific application.
The planned Phase 3 RESTORE-ALS study is relevant because it is intended to evaluate clinical outcomes more directly. Clene expects to begin the trial in the second quarter of 2027, with survival as the primary endpoint and time to clinical worsening events among the secondary efficacy measures. The programme is planned as a confirmatory study, although its final regulatory role will depend on the FDA’s decisions and the company’s ability to conduct it.
Can artificial intelligence improve regulatory analysis without compromising statistical integrity?
The use of artificial intelligence in clinical data processing creates opportunities for greater efficiency, particularly where tasks involve converting detailed specifications into standardised analytical outputs. However, pharmaceutical regulatory submissions require a level of reproducibility and traceability that differs from ordinary business reporting. Statistical results must be generated according to appropriate methods and remain open to independent verification.
Veristat’s InStat platform uses artificial intelligence-assisted specification development alongside validated statistical engines. This separation is intended to reduce the risk associated with allowing a language model to generate unverified statistical code directly. The resulting calculations still depend on the accuracy of the specifications and the integrity of the underlying data.
A critical requirement is that the system correctly identifies the intended analysis population, endpoint definition, statistical method and treatment comparison. If an incorrect specification is accepted, an automated system may reproduce the resulting error consistently. Automation therefore changes the location of quality-control risks rather than eliminating them.
Veristat has stated that its platform produces traceable outputs and that expert biostatisticians review the results. The company also reported favourable comparisons against traditional manual programming in its internal evaluations. However, those performance claims have not been independently established across a broad range of clinical programmes in the November announcement.
Regulatory acceptability depends on the complete quality system surrounding the analytical process. Sponsors must maintain appropriate data integrity, validation, documentation, access control and review procedures. These obligations remain in place regardless of whether outputs are generated manually or through an automated platform.
The broader opportunity is therefore not simply producing tables faster. It is developing a process that can reduce repetitive programming while maintaining reliable statistical execution and allowing qualified specialists to focus on interpretation and quality assurance. Demonstrating that balance across different study designs will be essential to wider adoption.
How does Clene’s financial position influence the significance of the Veristat partnership?
Clene’s third-quarter 2026 results reveal why operational efficiency is particularly relevant to its development strategy. The company ended September with approximately $9.9 million in cash and cash equivalents, compared with $5.2 million at the end of 2025. Management expects its available resources to support operations through approximately the middle of the first quarter of 2027.
Clene completed a $4 million registered direct equity offering in September, providing additional liquidity ahead of the FDA filing decision. The company has also indicated that acceptance of the CNM-Au8 application could create the potential for approximately $6.7 million to $7.8 million in proceeds from outstanding warrants issued earlier in 2026. Those proceeds are conditional and should not be treated as cash already received.
The company’s financial statements also contain a going-concern warning. Management disclosed substantial doubt about its ability to continue operations beyond the relevant assessment period without additional financing. This reflects the continuing funding requirements of a clinical-stage biotechnology company that has not yet established meaningful commercial revenue from an approved medicine.
For the quarter ended September 30, Clene reported research and development expenditure of approximately $4.4 million, up from $3.5 million a year earlier. The increase reflected factors including regulatory preparation, manufacturing work and personnel costs. General and administrative expenditure was approximately $1.9 million, compared with $2.2 million in the prior-year period.
Clene reported quarterly net income of approximately $3.9 million, but that accounting result was driven by non-cash changes in the fair value of warrant and derivative liabilities. It does not establish that the underlying biotechnology operations have become profitable. Operating losses, cash consumption and access to capital remain more meaningful indicators of financial sustainability.
The Veristat collaboration may have helped Clene avoid delays associated with the preparation of additional regulatory analyses. However, the announcement does not disclose the fees paid for the service or quantify the resulting financial savings. The reported 80% improvement concerns the analytical production timeline rather than a demonstrated 80% reduction in regulatory expenditure.
Clene’s funding position also creates a difficult sequencing issue. The company expects an FDA filing decision before the end of 2026, while its current cash runway extends only into early 2027. A favourable regulatory development could improve its financing options, but acceptance of an application would not guarantee approval or provide immediate commercial revenue.
What does the collaboration mean for clinical research organisations and pharmaceutical sponsors?
Contract research organisations increasingly compete on their ability to combine regulatory expertise, statistical analysis and technology-enabled execution. Pharmaceutical sponsors may outsource these functions when internal resources are limited or when specialised analytical requirements arise during development. Automation offers an opportunity to improve the efficiency of certain activities without requiring every sponsor to build equivalent systems internally.
Veristat’s reported experience with Clene provides an early commercial demonstration of its InStat platform. The company said it has observed similar performance in work involving a second sponsor, although the announcement does not disclose a comprehensive independent benchmark. Wider adoption will depend on reproducibility across different datasets, statistical methods and submission requirements.
For smaller biotechnology companies, the potential value may be especially significant when the timing of a regulatory submission affects financing, partnership discussions or development priorities. Reducing repetitive analytical programming could help sponsors respond more efficiently to changing data requirements. Nevertheless, submission speed is only one component of the drug development process.
Clinical trials must still generate suitable evidence, datasets must be complete and accurate, and regulatory authorities must determine whether the application supports the proposed indication. Faster production of tables cannot correct unsuccessful primary endpoints or establish a clinical benefit that the evidence does not demonstrate. The most useful role of automation is therefore to improve the preparation and verification of valid analyses.
The platform may also face implementation challenges. Sponsors use different electronic data capture systems, statistical programming conventions and quality-control procedures. A system intended for broad deployment must accommodate these variations while maintaining traceability and compliance.
Veristat describes InStat as system-agnostic, allowing sponsors to retain their preferred data capture environments and output formats. This flexibility may reduce implementation barriers, although each deployment still requires appropriate validation. Its long-term commercial success will depend on whether clients can demonstrate meaningful savings without compromising regulatory confidence.
What will determine whether CNM-Au8 advances beyond its submitted New Drug Application?
The immediate regulatory milestone is the FDA’s decision on whether to accept Clene’s October 2026 New Drug Application for substantive review. Acceptance would represent an important procedural step, but the regulator would still need to assess the complete clinical, biomarker and safety evidence. The submission’s central scientific challenge concerns whether the available information can support accelerated approval despite the original randomised trials not meeting their primary efficacy endpoints.
The planned RESTORE-ALS Phase 3 study provides a separate clinical development objective. Its intended focus on survival and clinical worsening is important because these outcomes are directly relevant to patients with amyotrophic lateral sclerosis. However, initiating and completing a large confirmatory trial will require significant resources, particularly given Clene’s limited reported cash runway.
For Veristat, the next commercial test is whether the InStat platform can reproduce its reported time savings across additional sponsors and more complex clinical programmes. The combination of automated specifications, validated statistical execution and expert oversight offers a credible workflow proposition, but its performance must be evaluated against the accuracy and traceability requirements of pharmaceutical regulation.
The collaboration demonstrates that Clene was able to accelerate an important component of its regulatory submission preparation. It does not reduce the evidentiary standard the FDA must apply to CNM-Au8 or eliminate the company’s financing requirements. The ultimate significance of the reported 80% improvement will depend on whether automation can deliver reliable regulatory analyses consistently, while Clene’s prospects remain tied to FDA acceptance, the strength of its clinical evidence and its ability to finance the next development stage.
