Search This Blog

Wednesday, August 20, 2025

5 Biotechs Taking the Pipeline-in-a-Product Approach to Drug Development

 

The platform strategy of using one molecule to target an underlying biological pathway to address many different diseases can be a goldmine for smaller companies. But it also has a unique set of challenges.

From 2012 to 2020, AbbVie’s I&I blockbuster Humira reigned as the most valuable drug in the world. Sales peaked in 2022, when the biologic brought in $21.2 billion and became the highest-grossing drug of all time. Now, after losing patent protection and market share to biosimilars, Humira has ceded the throne to Merck’s PD-1 inhibitor Keytruda, which last year hit upwards of $29 billion in revenue, comfortably beating all its competitors.

The market dominance of these two drugs was driven by several factors, but there is one crucial similarity between them: Both are what pharma analytics firm Evaluate calls pipelines-in-products. That is, drugs from which “several chronic disease indications can be squeezed out.”

In order for a molecule to be a good pipeline-in-a-product candidate it must target “fundamental biological processes . . . that underlie multiple disease states,” BioPharmCatalyst senior director John Gagliano told BioSpace in an email. Immune cascades, cellular checkpoints and metabolic pathways are good examples of these processes. There should also be a “clear biological rationale” that links the biological pathway the asset is targeting to the diseases it seeks to address, he added.

But broad mechanistic applicability is just the beginning. A good pipeline-in-a-product should also have “compelling clinical differentiation” as well as a “strong safety profile,” which in turn will allow its investigation across different populations, Gagliano explained.

If a molecule ticks all these boxes and clears all regulatory hurdles, it could be a big commercial boon. “In today’s therapeutic landscape, these products typically unlock the power of combination therapies,” Ken Krisko, partner and head of Life Sciences Corporate Partnering and Licensing at Cooley LLP, told BioSpace in an email.

The platform strategy also comes with “the potential for extended revenue over time,” Krisko continued, allowing companies to “place larger bets on the development and manufacturing infrastructure” for these molecules. If approved, he added, pipeline-in-a-product assets “offer unparalleled ability to combine commercialization strategy and improve leverage in payer negotiations.”

Of course, these are all only true if the company can execute in the market. And on that front, this platform approach comes with a unique set of barriers and challenges, especially for smaller companies, including the operational strain that comes with running drug development and commercialization in parallel.

Less-established players will need to “navigate a crowded competitive landscape, supporting each new indication while allocating resources strategically to avoid internal competition between launches,” according to Krisko.

“The key is recognizing that while these assets offer tremendous opportunity, they require sustained strategic commitment and excellent execution across multiple launches,” Gagliano told BioSpace.

In this piece, BioSpace reviews up-and-coming pipeline-in-product molecules, going beyond the Keytrudas and the Humiras—even past the GLP-1s—to focus on assets from smaller players that are taking on big risk for a huge potential reward.

Cantex Harnasses RAGE for Cancer, Respiratory

Cantex Pharmaceuticals is leveraging the pipeline strategy to address diseases across more than one therapeutic area. The Florida biotech is building its business on the small-molecule drug azeliragon, which according to CEO Stephen Marcus is “relevant to a broad range of illnesses.”

Azeliragon targets a central receptor called RAGE, which “has been strongly implicated in the initiation and progression” of many different common cancers and lung conditions, Marcus explained to BioSpace in an email. In line with this activity, Cantex is advancing the drug for pancreatic cancer, breast cancer and glioblastoma, as well as for brain metastasis from different primary malignancies. The FDA in December 2024 granted azeliragon orphan drug designation for brain metastasis from breast cancer.

Cantex is also working on respiratory indications, including severe asthma and advanced chronic obstructive pulmonary disease. Of these, the most mature program for azeliragon is in severe pneumonia. A Phase III study—which was originally in COVID-19 but later expanded to include all hospitalized pneumonia patients—will test whether azeliragon’s RAGE action can prevent acute kidney injury, a common complication that can lead to death in these patients. The study is ongoing and has a primary completion date in April next year.

But while azeliragon’s mechanism gives Cantex “multiple options for clinical development,” Marcus said that small companies like his still face the “risk that a lack of focus will produce insufficient evidence of effectiveness across” all of its potential indications.

“Careful selection of the disease targets based on likelihood of efficacy, the cost of clinical trials, and the need for such a therapy are crucial,” he added.

Cognition Focuses on Neuro With Zervimesine

Whereas Cantex is looking to apply its platform strategy across a diverse set of therapeutic areas, New York– and Pennsylvania-based Cognition Therapeutics is leaning into the neurodegenerative space.

Zervimesine is an orally available drug that works by targeting the sigma-2 receptor. This mechanism, according to CEO Lisa Ricciardi, is “functionally distinct from other approaches for neurodegenerative diseases” and allows zervimesine to “displace” toxic amyloid-beta plaques from the spaces between neurons, in turn preventing damage and potentially slowing cognitive decline.

Cognition is proposing the drug for Alzheimer’s disease, dementia with Lewy bodies and geographic atrophy secondary to age-related macular degeneration, with recent data underlining the therapeutic potential of zervimesine for all three conditions.

At the 2025 Alzheimer’s Association International Conference in June, results from the Phase II SHIMMER trial showed that in patients with dementia with Lewy bodies, zervimesine improved neuropsychiatric, cognitive, motor and functional performance. Also at AAIC 2025, the Phase II SHINE trial found that zervimesine could arrest further cognitive deterioration in patients with mild or moderate Alzheimer’s disease. Cognition has completed a “productive” end-of–Phase II meeting with the FDA for zervimesine in this indication.

Similarly, the mid-stage MAGNIFY trial documented slower lesion growth in patients with geographic atrophy. “To our knowledge, there is no other agent which has shown positive Phase II results across these disease entities,” Ricciardi told BioSpace in an email.

For smaller players like Cognition, Ricciardi said that pipelines-in-products such as zervimesine offer a “strategic advantage” by allowing for “resource optimization” while also giving the companies access to a “broadened market potential.”

Airway Focuses on the Lungs With Zelpultide Alpha

For Marc Salzberg, CEO and chief medical officer of Airway Therapeutics, a good pipeline-in-a-product must possess at least one of two key characteristics. It should either address underlying mechanisms for multiple conditions or have multiple properties. “In the case of zelpultide alpha, both apply,” Salzberg told BioSpace in an email.

Zelpultide alpha is a recombinant human protein that works by mimicking the endogenous protein SP-D, a lung surfactant that under healthy conditions helps clear pathogens from the lungs while also minimizing inflammation. This mechanism of action allows zelpultide alpha to modulate pulmonary inflammatory and immune responses, while also reducing bacterial, fungal and viral infections, according to Airway’s website.

The Ohio biotech is primarily testing zelpultide alpha for bronchopulmonary dysplasia (BPD) in infants, for which it launched a global Phase III study in December 2024. The trial aims to enroll more than 300 preterm infants who lack SP-D in their lungs. Phase Ib data, presented May 2024, showed that zelpultide alpha lowered BPD incidence in preterm neonates versus air sham, while also reducing the need for mechanical ventilation.

“It is essential to design a development program which is prioritized,” Salzberg explained to BioSpace, adding that small companies like Airway “cannot afford to develop a platform therapy across multiple indications in parallel.” In choosing BPD as its priority indication, Airway looked at the strength of the data backing zelpultide alpha, the level of medical need and the competitive landscape, he added.

Still, Airway plans to capitalize on zelpultide alpha’s broad potential for respiratory indications. The biotech is also testing the molecule for community-acquired pneumonia, for which it is in early-stage development. In addition, the company sees zelpultide alpha’s therapeutic potential for asthma, chronic obstructive pulmonary disease and acute respiratory distress syndrome, as per Airway’s website. The first two indications mirror another famed pipeline in a product, Regeneron and Sanofi’s blockbuster Dupixent.

Diakonos Chases Keytruda With a Dendritic Cell Vaccine

With Keytruda currently reigning as king of the biopharma industry, many smaller players are looking to follow in its footsteps and develop drugs that can be used across a wide variety of oncology indications.

One of these is Houston-based Diakonos Oncology, which is working on a dendritic cell vaccine called dubodencel. Part of the immune system, dendritic cells work by processing antigens—parts of a virus, for example, or bits of a tumor—to other immune players, in turn triggering an anti-cancer response. Diakonos is taking advantage of this pathway with a vaccine that uses a patient’s own cancer antigens to trigger a “powerful and natural immune response” against the cancer, according to the biotech’s website.

Diakonos is advancing dubodencel in a Phase II trial for glioblastoma, with the first patient dosed last month. Phase I data in June showed that vaccination led to significant expansion of CD4+ and CD8+ memory T cells, with side effects being mostly mild. At 12 months, the overall survival rate was higher than the standard of care.

For smaller companies like Diakonos, pipeline assets such as dubodencel “inevitably involve higher upfront complexity and costs,” CEO Jay Hartenbach told BioSpace in an email. In May, the biotech raised $20 million to support the development of dubodencel in glioblastoma, but even with this added infusion, Diakonos is approaching the drug’s development with discipline.

“Success goes hand in hand with disciplined sequencing of trials,” Hartenbach said, noting that by wisely planning trials, companies can “leverage strategic partnerships for specific indications,” such as through combo studies.

“Being able to target multiple indications does not mean the company should pursue them all at once,” he said.

Immunophotonics Targets Solid Tumors With Injectable Polymer

Like Diakonos, Missouri-based Immunophotonics is also chasing after Keytruda with IP-001, a glycan polymer that functions both as an antigen depot and an activator of the immune system. Owing to its structure, IP-001 helps retain antigens at the injection site and prolongs their exposure to immune cells. The molecule also attracts antigen-presenting cells, facilitating their activation, in turn triggering strong downstream immune responses.

“The versatility of IP-001 lies in its ability to be paired with various local treatments,” which in turn converts these procedures—such as photothermal therapy and tumor ablation—“into an immunologically active event,” Cooley’s Krisko told BioSpace. “One platform, multiple therapeutic avenues.”

Lu Alleruzzo, CEO and co-founder of Immunophotonics, told BioSpace in an email that platform assets like IP-001 need a development strategy that takes “expansion and alternate use into account.” The biotech is putting this into practice with its INJECTABL clinical program, which is testing IP-001 in various solid tumors, including colorectal cancer, soft tissue sarcoma and non-small cell lung cancer.

Last month, the company announced that the final patient had been dosed in INJECTABL-1, a Phase Ib/IIa trial in 41 patients with advanced disease. INJECTABL-3 will test IP-001 plus ablation treatments in certain key indications for which ablation is the current therapeutic standard, as per a September 2024 news release.

Small biotechs should also consider external partnerships, Alleruzzo added, for “effective execution of pipeline expansion.” Partners could also open up combination regimens. Immunophotonics is also running the Phase II/III INJECTABL-3 trial in collaboration with the Cardiovascular and Interventional Radiological Society of Europe, which will nominate a steering committee to help with the study design, as well as leverage its contract research organization Next Research.

https://www.biospace.com/business/5-biotechs-taking-the-pipeline-in-a-product-approach-to-drug-development

FDA Looks To Raise Survival Bar for Cancer Drug Approvals in New Draft Guidance

 

The FDA recommends that companies use overall survival as a primary endpoint for clinical trials where feasible. The new guidance follows the surprising return of CBER Head Vinay Prasad, who has previously argued for prioritizing OS.

The FDA wants to put a stronger emphasis on overall survival outcomes for the investigational cancer therapies it reviews, a move that could make it harder—but not impossible—for companies to seek accelerated approval for their products.

The new rules, outlined in a draft guidance published on Tuesday, come less than two weeks after the unexpected return of Center for Biologics Evaluation and Research Head Vinay Prasad, who has previously argued for prioritizing overall survival (OS).

The draft aims to guide drug sponsors on the use of survival data in their cancer drug applications, “with an emphasis on the analysis of overall survival as a pre-specified safety endpoint.”

When feasible, drug developers should prioritize using OS as the primary endpoint in their studies, according to the document. In cases where using OS as the primary efficacy outcome is not feasible—as in cancers that progress very slowly, for instance, or where treatments exist that are very effective and result in long survival times—companies should still collect OS data and submit them to the regulator.

“Overall survival is both an efficacy and a safety endpoint; it can be favorably impacted by the therapeutic benefits of a specific drug and negatively impacted by the drug’s toxicity,” the FDA wrote, noting that OS is the gold standard outcome in oncology. “Overall survival is also an objective, clinically meaningful endpoint that can be measured easily and precisely.”

To avoid confounding OS analysis, the FDA also recommends to “limit” the use of cross-over study designs, permitting these only in diseases with no or very limited therapeutic options.

“The heightened emphasis on OS - particularly in the context of analyzing safety of a novel drug/regimen - is not surprising, given the current FDA leadership,” analysts at Truist Securities wrote in a note to investors on Tuesday. The firm is also “encouraged” by the fact that accelerated approval still seems to be a possibility for companies despite the “stricter requirements” for OS in confirmatory follow-up.

For his part, Prasad has previously argued, in a paper published this March in Nature Reviews Clinical Oncology, that using surrogate endpoints for approval trades speed for therapeutic uncertainties for patients.

“If the FDA and other regulators were to shift their priority to patient-centered outcomes, such as OS, I argue that such a shift would probably lead to fewer, but also a higher standard of drugs entering the market,” he wrote.

Some key unanswered questions still remain, Truist continued. “We note that this draft guidance does not address accelerated approvals on the basis of single arm studies,” a practice that has grown increasingly common in recent years.

Tuesday’s draft document does dedicate a section to accelerated approval, which it says will be appropriate to use for drugs with “significant uncertainty” in OS findings but which have enough evidence to support efficacy via an intermediate clinical endpoint, such as treatment response or progression-free survival.

FDA guidelines are legally nonbinding and legally unenforceable—even when finalized. Companies are allowed to take an alternative approach to these recommendations, for which they should contact the agency.

https://www.biospace.com/fda/fda-looks-to-raise-survival-bar-for-cancer-drug-approvals-in-new-draft-guidance

Gabbard: Deep State Agents Are Sabotaging US Elections

 Via VigilantFox.com

Hannity just asked DNI Tulsi Gabbard whether there are still DEEP STATE actors inside the intel community SOBATAGING elections.

She didn’t even have to think about it, and said “Our national security depends” on exposing them.

Hannity: “Do we have deep state actors that are trying to influence our presidential elections? Is that what we are concluding here?”

Gabbard: “Yes.”

She said Brennan, Clapper, Comey, and their allies inside the agencies all worked to manipulate intelligence to serve partisan interests.

“These are bad actors that have to be rooted out.”

“Our national security depends on it. The ability for the American people’s trust to be earned back depends on exposing the bad actors and holding them accountable.”

“And that’s what President Trump is determined to do.”

https://www.zerohedge.com/political/gabbard-deep-state-agents-are-sabotaging-us-elections

Chinese Tech Giants Keep Buying Nvidia's H20 Chips Despite Beijing's Warnings

 Chinese demand for Nvidia’s H20 chips remains strong despite Beijing’s warnings over "security" risks, industry sources told Nikkei Asia. The U.S. chipmaker, which agreed to hand 15% of China sales revenue to Washington in exchange for export approval, is now eyeing a rebound in the world’s second-largest economy.

Beijing recently summoned Alibaba, ByteDance, Tencent, and Baidu, urging them to rely more on domestic suppliers like Huawei, Biren, and Cambricon. State media later echoed government concerns, alleging H20 chips carry backdoor risks — a claim Nvidia denies. Bloomberg also reported officials warning against using H20 in government projects.

Executives at major Chinese firms, however, say the guidance is only "advice," not a ban. One told Nikkei that their projects are not "critical infrastructure" and will continue to use Nvidia chips. Another noted that while Huawei’s Ascend is praised as "superior," it suffers from efficiency problems, higher-than-expected defects, and capacity constraints.

Tencent President Martin Lau recently said the company has enough GPUs for AI training and upgrades, but staff admitted concerns over shortages since Tencent also rents GPUs externally.

Price also works in Nvidia’s favor. "H20 is actually cheaper to use than some Chinese alternatives," a Beijing venture capitalist said, citing both purchase cost and energy efficiency. Nvidia’s CUDA software platform makes it harder to replace, even when Chinese rivals approach its hardware, according to Nikkei Asia.

A supplier executive said demand for H20 servers has not shifted since Beijing’s warning: "For the Chinese CSP [cloud service providers], they still hope to use higher-efficiency chips... they will continue to buy chips from Nvidia, but they will also increase their use of homegrown chips."

Margins on H20 are thinner, and production strains TSMC’s already tight CoWoS packaging capacity, but "being able to continue selling H20s into China is a way to show its goodwill to the Chinese government and hold on to its market share," the executive added.

Looking ahead, Nvidia is preparing a downgraded Blackwell chip for China, set to use GDDR7 memory instead of more advanced HBM, to comply with U.S. rules. Its performance is still unknown, and it’s unclear if Nvidia must again share revenue with Washington. Some hope the Trump administration will relax controls further.

"The H20 shows there are always deals to be made with the Trump administration," said the Beijing venture capitalist, suggesting even full-capacity Blackwell sales could be negotiated. But others doubt it. "I don't think there's the sense that they [Nvidia] necessarily need to" sell unrestricted Blackwell chips in China, said Bob O'Donnell of Technalysis Research.

Nikkei Asia writes that for now, Nvidia’s technology advantage keeps it indispensable. "Every company or country has recognized the importance of developing their own AI capabilities... they're going to buy Nvidia chips, whether they necessarily want to or not, because it's the only option they have," O'Donnell said.

"There's definitely politics going on, but I don't think it's going to be an issue [for Nvidia to recover in China]," added Dan Ives of Wedbush Securities.

Nvidia reiterated its chips are secure and warned that blocking H20 sales would harm U.S. interests.

https://www.zerohedge.com/markets/chinese-tech-giants-keep-buying-nvidias-h20-chips-despite-beijings-warnings

Ketamine and other NMDA receptor antagonists for chronic pain

 

Michael C Ferraro
Aidan G Cashin
Eric J Visser
Christina Abdel Shaheed
Michael A Wewege
Benedict M Wand
Sylvia M Gustin
Neil E O'Connella
James H McAuleya


Abstract


Rationale

N‐methyl‐D‐aspartate (NMDA) receptor antagonists are a group of medicines classed according to their mechanism of action. Ketamine and other NMDA receptor antagonists are used to treat chronic pain, despite uncertain benefits and harms.

Objectives

To evaluate the benefits and harms of ketamine and other NDMA receptor antagonists compared to placebo, usual care, or other medicines for adults with chronic non‐cancer, non‐headache pain.

Search methods

We searched CENTRAL, MEDLINE, Embase, and three trial registries (with reference checking, citation searching, and contact with study authors/experts) to identify included studies. The last search was 3 June 2025.

Eligibility criteria

We included randomised controlled trials (RCTs) in adults with chronic pain (≥ 3 months' duration), evaluating ketamine, memantine, dextromethorphan, amantadine, or magnesium versus placebo, usual care, or another medicine. We excluded studies of cancer or headache pain.

Outcomes

Critical outcomes were pain intensity and adverse events. Important outcomes were disability, depressive symptoms, health‐related quality of life, tolerability, and opioid consumption.

For adverse events and tolerability, follow‐up was until the end of treatment. For all other outcomes, we were interested in treatment effects in the immediate term (48 hours–1 week), short term (> 1 week–3 months), medium term (> 3 months–6 months), and long term (> 6 months).

Risk of bias

We assessed risk of bias using the Cochrane Risk of Bias tool for RCTs (RoB 2).

Synthesis methods

We converted all continuous pain intensity scores to a 0‐to‐100 scale (0 = no pain; 100 = worst pain). We synthesised results using random‐effects meta‐analysis where possible, reporting mean differences (MDs) for continuous outcomes and risk ratios (RRs) for dichotomous outcomes, each with its 95% confidence interval (CI). We assessed the certainty of evidence with GRADE.

Included studies

We found 67 RCTs (2309 participants): 30 parallel‐group RCTs (1568 participants) and 37 cross‐over RCTs (741 participants). Most studies (96%) were from high‐income countries. Female participation ranged from 11% to 100%. The interventions were ketamine (39 studies), memantine (10 studies), dextromethorphan (9 studies), amantadine (3 studies), and magnesium (8 studies). Sixty‐two studies used placebo comparators. Our quantitative synthesis included 28 studies.

Synthesis of results

Results are presented for pain intensity (continuous measures, at reported time points) and total adverse events.

Ketamine

Intravenous ketamine versus placebo

There is no clear evidence that intravenous ketamine reduces pain intensity in the immediate term (MD −15.79, 95% CI −32.09 to 0.51; 3 studies, 173 participants; very low certainty), short term (MD −5.32, 95% CI −15.51 to 4.87; 4 studies, 114 participants; low certainty), or medium term (MD −8.70, 95% CI −31.05 to 13.65; 1 study, 19 participants; very low certainty).

Intravenous ketamine may increase the risk of adverse events (RR 3.26, 95% CI 1.05 to 10.09; 4 studies, 140 participants; low certainty).

Oral ketamine versus placebo

There is no clear evidence that oral ketamine reduces pain intensity in the immediate term (MD −2.64, 95% CI −13.42 to 8.14; 2 studies, 46 participants; low certainty) or short term (MD −9.80, 95% CI −23.55 to 3.95; 2 studies, 40 participants; low certainty).

No studies reported total adverse events.

Topical ketamine versus placebo

There is no clear evidence that topical ketamine reduces pain intensity in the immediate term (MD 1.90, 95% CI −18.73 to 22.53; 1 study, 47 participants; very low certainty) or short term (MD 2.82, 95% CI −14.49 to 20.12; 2 studies, 64 participants; low certainty).

There is no clear evidence that topical ketamine increases the risk of adverse events (RR 1.14, 95% CI 0.47 to 2.73; 1 study, 47 participants; low certainty).

Memantine

Oral memantine versus placebo

There is no clear evidence that oral memantine reduces pain intensity in the immediate term (MD 4.00, 95% CI −9.93 to 17.93; 1 study, 36 participants; very low certainty), short term (MD −8.69, 95% CI −19.40 to 2.02; 6 studies, 217 participants; very low certainty), or medium term (MD −1.74, 95% CI −43.18 to 39.70; 2 studies, 101 participants; very low certainty).

There is no clear evidence that oral memantine increases the risk of adverse events (RR 1.09, 95% CI 0.76 to 1.56; 3 studies, 100 participants; low certainty).

Dextromethorphan

Oral dextromethorphan versus placebo

The evidence is very uncertain about the effect of oral dextromethorphan on pain intensity in the short term (MD −9.00, 95% CI −22.86 to 4.86; 1 study, 40 participants; very low certainty).

The evidence is very uncertain about the risk of adverse events with oral dextromethorphan (RR 1.80, 95% CI 0.73 to 4.43; 1 study, 40 participants; very low certainty).

Amantadine

Oral amantadine versus placebo

The evidence is very uncertain about the effect of oral amantadine on pain intensity in the immediate term (MD 6.00, 95% CI −12.45 to 24.45; 1 study, 26 participants; very low certainty).

The evidence is very uncertain about the risk of adverse events with oral amantadine (RR 0.86, 95% CI 0.14 to 5.20; 1 study, 26 participants; very low certainty).

Magnesium

Intravenous magnesium versus placebo

There is no clear evidence that intravenous magnesium reduces pain intensity in the immediate term (MD −2.00, 95% CI −14.43 to 10.43; 1 study, 55 participants; low certainty) and short term (MD −3.47, 95% CI −15.25 to 8.31; 2 studies, 82 participants; low certainty).

The evidence is very uncertain about the risk of adverse events with intravenous magnesium (0/35 events in intravenous magnesium group versus 0/35 in placebo group; 1 study, 70 participants; very low certainty).

Oral magnesium versus placebo

There is no clear evidence that oral magnesium reduces pain intensity in the short term (MD −0.55, 95% CI −8.32 to 7.21; 2 studies, 118 participants; low certainty).

No studies reported total adverse events.

Authors' conclusions

Limited low‐ to very low‐certainty evidence limits conclusions about the effects of ketamine, memantine, dextromethorphan, amantadine, and magnesium on pain intensity. Intravenous ketamine may increase the risk of adverse events, but the harms of ketamine and other NMDA receptor antagonists are generally unclear. Adequately powered RCTs are needed to determine the benefits and harms of ketamine and other NMDA receptor antagonists for chronic pain.

Funding

No dedicated funding.

Registration

Protocol available: doi.org/10.1002/14651858.CD015373

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD015373.pub2/full#CD015373-abs-0001


'Inner speech in motor cortex and implications for speech neuroprostheses'

Erin M. Kunz1,2,18,20 ekunz@stanford.eduBenyamin Abramovich Krasa3,18 ∙ Foram Kamdar4 ∙ … ∙ Shaul Druckmann1,17 ∙ Jaimie M. Henderson2,4 ∙ Francis R. Willett

Highlights

Attempted, inner, and perceived speech have a shared representation in motor cortex
An inner-speech BCI decodes general sentences with improved user experience
Aspects of private inner speech can be decoded during cognitive tasks like counting
High-fidelity solutions can prevent a speech BCI from decoding private inner speech

Summary

Speech brain-computer interfaces (BCIs) show promise in restoring communication to people with paralysis but have also prompted discussions regarding their potential to decode private inner speech. Separately, inner speech may be a way to bypass the current approach of requiring speech BCI users to physically attempt speech, which is fatiguing and can slow communication. Using multi-unit recordings from four participants, we found that inner speech is robustly represented in the motor cortex and that imagined sentences can be decoded in real time. The representation of inner speech was highly correlated with attempted speech, though we also identified a neural “motor-intent” dimension that differentiates the two. We investigated the possibility of decoding private inner speech and found that some aspects of free-form inner speech could be decoded during sequence recall and counting tasks. Finally, we demonstrate high-fidelity strategies that prevent speech BCIs from unintentionally decoding private inner speech.





Layer-specific changes in sensory cortex across the lifespan in mice and humans